Methods for optical character recognition (OCR)

Abstract

A method is provided for Optical Character Recognition (OCR). A plurality of OCR decoding results each having a plurality of positions is obtained from capturing and decoding a plurality of images of the same one or more OCR characters. A recognized character in each OCR decoding result is compared with the recognized character that occupies an identical position in each of the other OCR decoding results. A number of occurrences that each particular recognized character occupies the identical position in the plurality of OCR decoding results is calculated. An individual confidence score is assigned to each particular recognized character based on the number of occurrences, with a highest individual confidence score assigned to a particular recognized character having the greatest number of occurrences. Determining which particular recognized character has been assigned the highest individual confidence score determines which particular recognized character comprises a presumptively valid character for the identical position.

Description

FIELD OF THE INVENTION

The present invention relates to reducing optical character recognition errors, and more particularly, relates to methods for optical character recognition (OCR).

BACKGROUND

Optical character recognition (referred to herein as OCR) is a useful feature that allows a computing device to recognize text (more particularly, characters thereof) in an image and convert the text of the image into machine-operable text, e.g., ASCII characters. The machine-operable text is considered an “OCR decoding result”. “Machine-operable text” includes text characters that can be processed in a computer, usually as bytes. For example, users can download, photograph, or scan books, documents, product labels, etc. to obtain an image including text. The users can perform OCR on the image so as to recognize the text in the image, thereby allowing a user on his/her computer, mobile phone, tablet, etc. to select, copy, search, and edit the text.

Conventional OCR systems, however, frequently produce OCR errors (referred to as “misreads”) when recognizing and decoding text. Common errors include unrecognizable or improperly converted text, e.g., an “O” (letter O) for a “0” (number zero), or an “E” for a “B”. OCR errors can often render converted text unusable until a user corrects the errors. Improperly converted text can occur, for instance, when an image has a low resolution, blurred text, and/or unclear text. In another instance, conventional OCR systems may improperly convert text because the image may include uncommon characters or an underlying adjacent graphic that obscures the text. Furthermore, OCR systems can recognize illustrations in an image as text when the illustration does not actually include text. Generally speaking, Optical Character Recognition (OCR) has conventionally had an unacceptable misread rate.

In barcode scanning, a voting methodology is used to compare the decoded data string to subsequent decodes and when a sufficient number of identical scans occur, the decode is presumed to be valid and passed onto the host or application. Misreads are rare in barcode scanning and customers expect the same from OCR scanning. Unfortunately, the voting methodology does not work well in optical character recognition due to the frequency of OCR misreads—it may take a lot of scanning before a sufficient number of identical scans occur without a misread, if at all.

With an expected decoding result, a standard computer program may be conventionally used to identify a misread OCR decoding result by identifying a misread character(s) therein. For example, a plurality of OCR decoding results are depicted below, with multiple misreads detected by the standard computer program (for example, in the first line, “THEEASYDOG” should be “THELAZYDOG”):

OCR Line 3: P<UTOTHEEAZYDOG<<QUICK<BROWN<FOX<JUMPS<OVER<MISREAD

OCR Line 3: P<UTOTHELAZYDOG<<QUICK<BROWN<FOX<JUMPS<OVER<MISREAD

OCR Line 3: P<UTOTHELAZYDOG<<QUICK<BROWN<FOX<JUMPS<OVER<OK

OCR Line 3: P<OTOTHELAZYDOG<<QUICK<BROWN<FOX<JUMPS<OVER<MISREAD

OCR Line 3: P<UTOTHELAZYDOG<<QUICK<BROWN<FOX<JUMPS<OVER<MISREAD

OCR Line 3: P<UTOTHELAZYDOG<<QUICK<BROWN<FOX<JUMPS<OVER<MISREAD

OCR Line 3: P<UTOTHBLAZYDOG<<QUICK<BROWN<FOX<JUMPS<OVER<MISREAD

OCR Line 3: P<UTOTHELAZYDOG<<QUICK<BROWN<FOX<JUMPS<OVER<MISREAD

OCR Line 3: P<UTOTHELAZYDOG<<QUICK<BROWN<FOX<JUMPS<OVER<MISREAD

OCR Line 3: P<UTOTHELAZYDOG<<QUICK<BROWN<FOX<JUMPS<OVER<MISREAD

However, in the usual application in the field in which the decoding result is unknown in advance, the standard computer program is not useable for detecting a misread, and therefore for determining when a valid OCR decoding result has been obtained.

Therefore, a need exists for methods for optical character recognition (OCR). Various embodiments provide a presumptively valid OCR decoding result with a high level of confidence even if every OCR decoding result contains a misread.

SUMMARY

Accordingly, in one aspect, the present invention embraces a method for Optical Character Recognition. A plurality of OCR decoding results each having a plurality of positions is obtained from capturing and decoding a plurality of images of the same one or more OCR characters. A recognized character in each OCR decoding result is compared with the recognized character that occupies an identical position in each of the other OCR decoding results. A number of occurrences that each particular recognized character occupies the identical position in the plurality of OCR decoding results is calculated. An individual confidence score is assigned to each particular recognized character based on the number of occurrences, with a highest individual confidence score assigned to a particular recognized character having the greatest number of occurrences. Determining which particular recognized character has been assigned the highest individual confidence score determines which particular recognized character comprises a presumptively valid character for the identical position.

In another aspect, the present invention embraces a method for Optical Character Recognition (OCR). The method comprises obtaining a plurality of OCR decoding results from capturing and decoding a plurality of images of the same one or more characters. Each OCR decoding result comprises a plurality of positions. A first recognized character of each OCR decoding result of the plurality of OCR decoding results is compared with the first recognized character of each of the other OCR decoding results to determine one or more first recognized characters. Each first recognized character of the one or more first recognized characters is assigned an individual confidence score according to its respective number of occurrences as the first recognized character in the plurality of OCR decoding results. Which particular first recognized character comprises a presumptively valid first recognized character is determined by determining the particular first recognized character that has been assigned a highest individual confidence score.

In another aspect, the present invention embraces a method for Optical Character Recognition (OCR). The method comprises obtaining a plurality of OCR decoding results from capturing and decoding a plurality of images of the same one or more characters. Each OCR decoding result comprises a plurality of positions. Each recognized character of the plurality of OCR decoding results is associated with an individual confidence score. A presumptively valid OCR decoding result is identified from its total confidence score based on a combination of the individual confidence scores assigned to each of the recognized characters thereof, the presumptively valid OCR decoding result having a highest total confidence score.

The foregoing illustrative summary, as well as other exemplary objectives and/or advantages of the present invention, and the manner in which the same are accomplished, are further explained within the following detailed description and its accompanying drawings.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG. 1 is a schematic diagram of an exemplary OCR scanner useful for obtaining a plurality of OCR decoding results (such as shown, for example, in the Background and in FIG. 3 below);

FIG. 2 is a flow diagram of a method for optical character recognition (OCR), according to various embodiments of the present invention; and

FIG. 3 graphically depicts a plurality of OCR decoding results, illustrating that methods for determining a presumptively valid OCR decoding result according to various embodiments of the present invention enable determination of a presumptively valid OCR decoding result even though every line of text (decoding result) contains a misread, according to various embodiments of the present invention.

DETAILED DESCRIPTION

Various embodiments are directed to methods for optical character recognition (OCR). Various embodiments reduce misreads and enable misread recognition so that decoding can be aborted. Various embodiments produce dependable OCR decoding results, even if none of the OCR decoding results from capturing and scanning the same line of text and even if none of the individual OCR decoding results is a valid decoding result. Various embodiments enable or generate a presumptively valid OCR decoding result with a high level of confidence in the decoding result. The confidence level may be arbitrarily set.

As used herein, the term “presumptively valid decoding result” refers to a decoding result that may or may not be valid, but is presumed valid because each recognized character in each position of the decoding result has been assigned the highest individual confidence score or a score greater than a minimum threshold.

Various embodiments of the present invention will be described in relation to a hand-held OCR scanner. However, the present invention may be equally applicable to other types and styles of OCR-scanners (e.g., a “page scanner”, such as the type used in Customs (e.g., a flat window upon which the official places a passport face down, the type where the traveler inserts the passport into a slot and the entire image is read, and the type whereby the user physically “swipes” the passport through the reader similar to the way a retailer reads a magnetic stripe)). As used herein, the term “OCR scanner” refers to a device that converts images of typed, handwritten or printed text into machine-encoded text, whether from a scanned document, a photo of a document, an image scan of a product at a point of sale (POS), a scene-photo (for example the text on signs and billboards in a landscape photo) or from subtitle text superimposed on an image (for example from a television broadcast), etc.

Referring now to FIG. 1, an exemplary OCR scanner 10 is depicted. The OCR scanner 10 could be a single purposed device dedicated to OCR scanning, an image based bar code scanner (i.e., imager) such as an imager equipped mobile computer with OCR capability, a linear imager or CCD, or could be a general purpose mobile computing device (MCD) configured by software to scan text using an integrated camera. The exemplary OCR scanner 10 includes a sensor 11 for sensing one or more characters in a line of text 4. The sensor 11 may use a variety of techniques to sense the one or more characters. The sensor may include a laser scanner for scanning a laser across a field of view 6. A collimated beam of laser light (e.g., 630-680 nanometer wavelength) may be swept back and forth along a scan-line 5 aligned with the line of text. A light detector converts the reflected light into a scanned-OCR signal. The scanned-OCR signal may be an electronic signal with a modulated amplitude corresponding to the pattern. This modulated signal may be converted into decoded data by a processor 12 (e.g., one or more controller, digital signal processor (DSP), application specific integrated circuit (ASIC), programmable gate array (PGA), a multi-core processor, and/or programmable logic controller (PLC)) communicatively coupled to the sensor 11.

In various embodiments, the sensor 11 includes an image sensor (e.g., CCD, CMOS sensor, etc.) for capturing images of a field of view 6. To scan a line of text, the field of view 6 of the scanner is positioned to view the line of text and an image is captured with the image sensor. The processor 12 communicatively coupled to the sensor 11 converts the image of the line of text into decoded data (a decoding result).

The exemplary OCR scanner 10 also includes a memory 16 (e.g., read-only memory (ROM), flash memory, a hard-drive, etc.) that stores information. The stored information may include a processor-executable software program for decoding the line of text 4. The processor 12 may access the memory 16 to execute the steps of a decoding program for decoding the line of text.

The decoding software program configures the processor 12 to receive the information from the sensor 11 and convert the scanned-text into a decoding result. In various embodiments, the OCR scanner includes an input/output (I/O) module. The I/O module 13 (e.g., user interface) may display the decoded information (e.g., visually and/or auditory).

The subsystems in the scanner 10 are electrically connected via a coupler (e.g., wires, traces, etc.) to form an interconnection subsystem 15. The interconnection system 15 may include power buses or lines, data buses, instruction buses, address buses, etc., which allow operation of the modules/subsystems and the interaction there between.

The scanner 10 is communicatively connected to a computer network 20 via a wired or wireless data link 19 (e.g., IEEE 802.11). A host computer 21 is also communicatively coupled to the computer network 20. This data link 19 may be accessed by a communication module 17 integrated with the scanner 10. In a wireless configuration, the communication module may communicate with a host device over the network via a variety of communication protocols (e.g., WI-FI®, BLUETOOTH®, CDMA, TDMA, or GSM). In some embodiments, the scanner 10 may incorporate a cellular telephone module to communicate over a cellular network as described in U.S. Pat. No. 6,212,401, which is incorporated in its entirety herein by reference.

Referring now to FIGS. 2 and 3, according to various embodiments of the present invention, a method 100 for Optical Character Recognition (OCR) is disclosed. The method 100 for Optical Character Recognition (OCR) comprises obtaining a plurality of OCR decoding results from capturing and decoding a plurality of images of the same one or more OCR characters (step 200). While the OCR decoding results of FIG. 3 use an OCR font, it is to be understood that OCR decoding results may use other than the OCR font shown, e.g., Arial, Times New Roman, Courier. Obtaining the plurality of OCR decoding results comprises capturing the plurality of images of the same one or more OCR characters and decoding each image after the plurality of images is captured or decoding after each image of the plurality of images is captured. FIG. 3 depicts ten OCR decoding results, each comprising a plurality of positions.

Still referring to FIGS. 2 and 3, according to various embodiments of the present invention, method 100 for Optical Character Recognition (OCR) comprises comparing a recognized character in each OCR decoding result of the plurality of OCR decoding results with the recognized character that occupies an identical position in each of the other OCR decoding results (step 210). For example, a first recognized character of each OCR decoding result of the plurality of OCR decoding results may be compared with the first recognized character of each of the other OCR decoding results to determine one or more first recognized characters. Each succeeding recognized character of each OCR decoding result can be compared with a recognized character occupying the identical position in each of the other OCR decoding results. In the depicted embodiment of FIG. 3 in which there are an exemplary forty-four positions, in the first decoding result, the letter “P” occupies the first position, the character < occupies the second position, the letter “U” occupies the third position, the letter “T” occupies the fourth position, the letter “O” occupies the fifth position, the letter “T” occupies the sixth position, the letter “H” occupies the seventh position, the letter “E” occupies the eighth position, the letter “E” (incorrectly) occupies the ninth position, and so on.

Still referring to FIG. 3, in accordance with various embodiments of the present invention, the letter “P” in the first position in the first decoding result is compared with the recognized character that occupies the identical position (the first position, in this case) in each of the other OCR decoding results. The character “<” in the second position in the first decoding result is compared with the recognized character that occupies the second position in each of the other OCR decoding results. The letter “U” in the third position in the first decoding result is compared with the recognized character that occupies the third position in each of the other OCR decoding results. The letter “T” in the fourth position in the first decoding result is compared with the recognized character that occupies the fourth position in each of the other OCR decoding results. As there are 44 exemplary positions in each of the depicted OCR decoding results of FIG. 3, the comparison step is repeated 44 times in the depicted embodiment to determine the presumptively valid character for each position as hereinafter described.

Still referring to FIGS. 2 and 3, according to various embodiments of the present invention, method 100 for Optical Character Recognition (OCR) comprises calculating a number of occurrences that each particular recognized character occupies the identical position in the plurality of OCR decoding results (step 220). For example, the letter “P” appears as the recognized character in the first position of the line of text in all ten decoding results (i.e., the letter “P” has ten occurrences (occ) and there is no other recognized character in the first position). Therefore, the letter “P” has the highest individual confidence score for the first position and is therefore the presumptively valid character for the first position. The character “<” appears as the recognized character in the second position of the line of text in all ten decoding results (i.e., the character “<” has ten occurrences and there is no other recognized character in the second position). Therefore, the character “<” has the highest individual confidence score for the second position and is therefore the presumptively valid character for the second position. Looking now at the third position in each decoding result of the ten exemplary decoding results, there are nine occurrences of the letter “U” and one occurrence of the letter “O”. The letter “U” in the third position, and not the letter “O”, is the presumptively valid character with more occurrences thereof in the third position relative to the number of occurrences of the letter “O”. There are ten occurrences of the letter “T” in the fourth position, ten occurrences of the letter “O” in the fifth position, ten occurrences of the letter “T” in the sixth position, and ten occurrences of the letter “H” in the seventh position. These recognized characters are therefore presumptively valid characters for each of their respective positions. However, in the eighth position, the letter “E” was misread as “B” in the seventh decoding result and in the ninth position, the letter “L” was misread as the letter “E” in the first OCR decoding result. For the eighth position, the letter “E” is the presumptively valid character with nine occurrences versus one occurrence for the letter “B”. For the ninth position, the letter “L” is the presumptively valid character with nine occurrences, whereas the letter “E” only has one occurrence. As noted previously, each succeeding recognized character of each OCR decoding result can be compared with a recognized character occupying the identical position in each of the other OCR decoding results. The next discrepancy between recognized characters in a respective position is in the letter “O” in DOG (i.e., the 14th position), with the fourth decoding result from the top misreading the number “0” instead of the letter “O”. As the letter “O” occupying the 14th position in nine out of ten decoding results (nine occurrences) versus the number “0” with one occurrence, the letter “O” is the presumptively valid character for the 14th position. In the word “BROWN”, the letter “B” occupies the 24th position in nine decoding results with the letter “E” occupying the 24th position for one occurrence. Therefore, for the 24th position, B is the presumptively valid character. The letter “O” in the word “BROWN” occupies the 26th position in each decoding result seven times out of ten decoding results and the number “0” has only three occurrences in the 27th position. Therefore, the letter “O” is the presumptively valid character for the 27th position. In the word “FOX”, the letter “O” has eight occurrences whereas the number “0” has two occurrences in the 31st position. Therefore, for the 31st position, the letter “O” is the presumptively valid character. In the word “JUMPS”, the letter “P” has nine occurrences whereas the letter “F” (third OCR decoding result of FIG. 3) has only one occurrence. Therefore, the letter “P” is the presumptively valid character for the 37th position as it has more occurrences than the letter “F” for that position. In the word “OVER”, the letter “O” in the 40^nd position has eight occurrences whereas the number “0” has only two occurrences, making the letter “O” the presumptively valid character for that position. The number of occurrences that a recognized character occupies a particular position in the plurality of OCR decoding results is calculated.

Stated simply, if the same recognized character occupies the same position in all decoding results, the recognized character is the presumptively valid character for that position (e.g., if the letter “A” occupies the same position in all decoding results, the letter “A” is the presumptively valid character for that position). On the other hand, if there is more than one recognized character occupying the same position in the plurality of decoding results, the presumptively valid character is the recognized character that occupies the same position in a greater number of decoding results (the recognized character with the greater number of occurrences) as hereinafter described.

The recognized character from capturing and decoding each image of the plurality of images may be transmitted to a memory space and each occurrence of the recognized character aggregated in the memory space for calculating the number of occurrences as previously described.

A lexicon for the plurality of OCR decoding results may be predetermined. If the recognized character is not included in the predetermined lexicon, it can be determined whether to include or exclude the recognized character in the calculation of the number of occurrences. For example, the number “0” is often misread as the letter “O”, and vice versa. By including the number “0” in the predetermined lexicon, i.e., it occurs in a numeric-only field such as a passport number, it may be included in the number of occurrences for the letter “O” so as to not over exclude. By excluding the number “0” in the predetermined lexicon, the number “0” could not be a presumptively valid character. Relatedly, different characters can have different levels of confidence depending on their propensity for misreading. Sometimes there are known attributes in a given field that can be applied. For example, for the decoding result “BROWN”, it is known that it is a non-numeric field so that misreads into a digit are not counted.

In various embodiments of the present invention, the method 100 for Optical Character Recognition (OCR) may further comprise verifying that the number of OCR decoding results in which the recognized character occupies the particular position comprises a minimum threshold number of OCR decoding results (step 230). If the calculated number of OCR decoding results in which the recognized character occupies the particular position does not comprise the minimum threshold number of OCR decoding results, additional OCR decoding results may be obtained by capturing and decoding additional images of the same one or more characters (step 240). The comparing, calculating, and assigning steps may then be repeated with the original OCR decoding results and the additional OCR decoding results until the calculated number of OCR decoding results in which the recognized character occupies the particular position comprises the minimum threshold number of OCR decoding result out of the total number of OCR decoding results (a minimum threshold percentage). For example, if a particular recognized character occupies a particular position in only half the decoding results, there may not be confidence that the particular recognized character is the presumptively valid character for that position. The minimum threshold number or percentage can be set to any threshold desired, for example, 7 in 10 or better.

If there is significant background noise or an out of focus situation, it may be possible that no one recognized character emerges as the recognized character with the highest number of occurrences and therefore no one recognized character can be assigned the highest individual confidence score. It may be necessary to obtain additional OCR decoding results (hopefully after eliminating at least some of the background noise or improving the focus). In this situation, the imager can be configured to capture and buffer images continually or the scanner can offer auditory or visual cues to the user to continue scanning.

Still referring to FIGS. 2 and 3, according to various embodiments of the present invention, the method 100 for Optical Character Recognition (OCR) comprises assigning an individual confidence score to each particular recognized character based on its respective number of occurrences, with a highest individual confidence score assigned to a particular recognized character having the greatest number of occurrences for that position (step 250) (i.e., the highest individual confidence score is assigned to a particular recognized character that occupies a particular position within the greatest number of OCR decoding results) (whether that includes original OCR decoding results, or original and additional OCR decoding results). The comparison, calculation, and assignment steps are completed over all recognized characters included in each position of each OCR decoding result of the plurality of OCR decoding results so that each particular recognized character has an individual confidence score.

Still referring to FIGS. 2 and 3, according to various embodiments of the present invention, the method 100 for Optical Character Recognition (OCR) comprises determining which particular recognized character comprises a presumptively valid character for each position by determining which particular recognized character for that position has been assigned the highest individual confidence score (step 260). As noted previously, if only one recognized character occupies the same position in each decoding result, that one particular recognized character has a 100% confidence score that it is the presumptively valid character for that particular position. While only two alternative recognized characters for some of the positions of FIG. 3 are described, it is to be understood that there may be multiple recognized characters in each position, with each occupying the position for a number of occurrences. For example, the 8th position could have had 7 E's, 2 B's, and an F. The recognized character occupying the same position the greatest number of occurrences is the presumptively valid character for that position. If there is a tie in the number of occurrences, it may be necessary to obtain additional decoding results until a particular recognized character leads by a selected margin as the recognized character in that particular position with the highest confidence score.

The method 100 for Optical Character Recognition (OCR) comprises repeating the comparing, calculating, assigning, and determining steps for each position in the decoding result to determine the presumptively valid character for each position. In various embodiments of the present invention, the comparing, calculating, assigning, and determining steps may be completed for each position prior to moving onto the next position where comparison, calculation, assignment, and determination steps are repeated for determining the presumptively valid character for the next position, and so on for each position. However, it is not necessary that all steps be completed in sequence prior to moving onto a next position. It may be possible to return to a position for determining the presumptively valid decoding result.

Still referring to FIGS. 2 and 3, according to various embodiments of the present invention, the method 100 for Optical Character Recognition (OCR) further comprises associating each OCR decoding result with a total confidence score based on a combination of the individual confidence scores assigned to each particular recognized character therein (step 270), identifying the OCR decoding result(s) that is/are associated with the highest total confidence score (step 280), and selecting the OCR decoding result(s) that is/are associated with the highest total confidence score as the presumptively valid OCR decoding result (step 290). There may be more than one presumptively valid OCR decoding result in the plurality of OCR decoding results obtained in step 200, as long as the presumptively valid OCR decoding results are identical.

The OCR decoding result may be selected as the presumptively valid decoding result even when the selected OCR decoding result does not match another OCR decoding result of the plurality of OCR decoding results or when none of the plurality of decoding results is correctly decoded in its entirety because the presumptively valid decoding result is the decoding result with the highest total confidence score.

In various embodiments of the present invention, it is to be understood that the presumptively valid decoding result may alternatively be generated from an ordered combination of the presumptively valid characters (the characters with the highest individual confidence scores), in which case the presumptively valid decoding result may be absent from the plurality of decoding results obtained in step 200 (step 300).

FIG. 3 demonstrates the improvement over the voting methodology used in barcode scanning. As noted previously, voting compares the data string to subsequent decodes and when a sufficient number of identical scans occur, the decode is passed. In accordance with various embodiments of the present invention, every recognized character in the decoding results is analyzed statistically and the decoding result may be a presumptively valid decoding result even if none of the scans matches a subsequent scan.

From the foregoing, it is to be understood that various embodiments determine a presumptively valid optical character recognition (OCR) decoding result. Various embodiments reduce misreads and enable misread recognition so that decoding can be aborted. Various embodiments help determine presumptively valid OCR decoding results, even if none of the individual OCR decoding results obtained is a valid decoding result. Various embodiments enable or generate a presumptively valid OCR decoding result with a high level of confidence in the decoding result.

To supplement the present disclosure, this application incorporates entirely by reference the following commonly assigned patents, patent application publications, and patent applications:

U.S. Pat. Nos. 6,832,725; 7,128,266; 7,159,783; 7,413,127; 7,726,575; 8,294,969; 8,317,105; 8,322,622; 8,366,005; 8,371,507; 8,376,233; 8,381,979; 8,390,909; 8,408,464; 8,408,468; 8,408,469; 8,424,768; 8,448,863; 8,457,013; 8,459,557; 8,469,272; 8,474,712; 8,479,992; 8,490,877; 8,517,271; 8,523,076; 8,528,818; 8,544,737; 8,548,242; 8,548,420; 8,550,335; 8,550,354; 8,550,357; 8,556,174; 8,556,176; 8,556,177; 8,559,767; 8,599,957; 8,561,895; 8,561,903; 8,561,905; 8,565,107; 8,571,307; 8,579,200; 8,583,924; 8,584,945; 8,587,595; 8,587,697; 8,588,869; 8,590,789; 8,596,539; 8,596,542; 8,596,543; 8,599,271; 8,599,957; 8,600,158; 8,600,167; 8,602,309; 8,608,053; 8,608,071; 8,611,309; 8,615,487; 8,616,454; 8,621,123; 8,622,303; 8,628,013; 8,628,015; 8,628,016; 8,629,926; 8,630,491; 8,635,309; 8,636,200; 8,636,212; 8,636,215; 8,636,224; 8,638,806; 8,640,958; 8,640,960; 8,643,717; 8,646,692; 8,646,694; 8,657,200; 8,659,397; 8,668,149; 8,678,285; 8,678,286; 8,682,077; 8,687,282; 8,692,927; 8,695,880; 8,698,949; 8,717,494; 8,717,494; 8,720,783; 8,723,804; 8,723,904; 8,727,223; 8,740,082; 8,740,085; 8,746,563; 8,750,445; 8,752,766; 8,756,059; 8,757,495; 8,760,563; 8,763,909; 8,777,108; 8,777,109; 8,779,898; 8,781,520; 8,783,573; 8,789,757; 8,789,758; 8,789,759; 8,794,520; 8,794,522; 8,794,525; 8,794,526; 8,798,367; 8,807,431; 8,807,432; 8,820,630; 8,822,848; 8,824,692; 8,824,696; 8,842,849; 8,844,822; 8,844,823; 8,849,019; 8,851,383; 8,854,633; 8,866,963; 8,868,421; 8,868,519; 8,868,802; 8,868,803; 8,870,074; 8,879,639; 8,880,426; 8,881,983; 8,881,987; 8,903,172; 8,908,995; 8,910,870; 8,910,875; 8,914,290; 8,914,788; 8,915,439; 8,915,444; 8,916,789; 8,918,250; 8,918,564; 8,925,818; 8,939,374; 8,942,480; 8,944,313; 8,944,327; 8,944,332; 8,950,678; 8,967,468; 8,971,346; 8,976,030; 8,976,368; 8,978,981; 8,978,983; 8,978,984; 8,985,456; 8,985,457; 8,985,459; 8,985,461; 8,988,578; 8,988,590; 8,991,704; 8,996,194; 8,996,384; 9,002,641; 9,007,368; 9,010,641; 9,015,513; 9,016,576; 9,022,288; 9,030,964; 9,033,240; 9,033,242; 9,036,054; 9,037,344; 9,038,911; 9,038,915; 9,047,098; 9,047,359; 9,047,420; 9,047,525; 9,047,531; 9,053,055; 9,053,378; 9,053,380; 9,058,526; 9,064,165; 9,064,165; 9,064,167; 9,064,168; 9,064,254; 9,066,032; 9,070,032; 9,076,459; 9,079,423; 9,080,856; 9,082,023; 9,082,031; 9,084,032; 9,087,250; 9,092,681; 9,092,682; 9,092,683; 9,093,141; 9,098,763; 9,104,929; 9,104,934; 9,107,484; 9,111,159; 9,111,166; 9,135,483; 9,137,009; 9,141,839; 9,147,096; 9,148,474; 9,158,000; 9,158,340; 9,158,953; 9,159,059; 9,165,174; 9,171,543; 9,183,425; 9,189,669; 9,195,844; 9,202,458; 9,208,366; 9,208,367; 9,219,836; 9,224,024; 9,224,027; 9,230,140; 9,235,553; 9,239,950; 9,245,492; 9,248,640; 9,250,652; 9,250,712; 9,251,411; 9,258,033; 9,262,633; 9,262,660; 9,262,662; 9,269,036; 9,270,782; 9,274,812; 9,275,388; 9,277,668; 9,280,693; 9,286,496; 9,298,964; 9,301,427; 9,313,377; 9,317,037; 9,319,548; 9,342,723; 9,361,882; 9,365,381; 9,373,018; 9,375,945; 9,378,403; 9,383,848; 9,384,374; 9,390,304; 9,390,596; 9,411,386; 9,412,242; 9,418,269; 9,418,270; 9,465,967; 9,423,318; 9,424,454; 9,436,860; 9,443,123; 9,443,222; 9,454,689; 9,464,885; 9,465,967; 9,478,983; 9,481,186; 9,487,113; 9,488,986; 9,489,782; 9,490,540; 9,491,729; 9,497,092; 9,507,974; 9,519,814; 9,521,331; 9,530,038; 9,572,901; 9,558,386; 9,606,581; 9,646,189; 9,646,191; 9,652,648; 9,652,653; 9,656,487; 9,659,198; 9,680,282; 9,697,401; 9,701,140; U.S. Design Pat. No. D702,237; U.S. Design Pat. No. D716,285; U.S. Design Pat. No. D723,560; U.S. Design Pat. No. D730,357; U.S. Design Pat. No. D730,901; U.S. Design Pat. No. D730,902; U.S. Design Pat. No. D734,339; U.S. Design Pat. No. D737,321; U.S. Design Pat. No. D754,205; U.S. Design Pat. No. D754,206; U.S. Design Pat. No. D757,009; U.S. Design Pat. No. D760,719; U.S. Design Pat. No. D762,604; U.S. Design Pat. No. D766,244; U.S. Design Pat. No. D777,166; U.S. Design Pat. No. D771,631; U.S. Design Pat. No. D783,601; U.S. Design Pat. No. D785,617; U.S. Design Pat. No. D785,636; U.S. Design Pat. No. D790,505; U.S. Design Pat. No. D790,546; International Publication No. 2013/163789; U.S. Patent Application Publication No. 2008/0185432; U.S. Patent Application Publication No. 2009/0134221; U.S. Patent Application Publication No. 2010/0177080; U.S. Patent Application Publication No. 2010/0177076; U.S. Patent Application Publication No. 2010/0177707; U.S. Patent Application Publication No. 2010/0177749; U.S. Patent Application Publication No. 2010/0265880; U.S. Patent Application Publication No. 2011/0202554; U.S. Patent Application Publication No. 2012/0111946; U.S. Patent Application Publication No. 2012/0168511; U.S. Patent Application Publication No. 2012/0168512; U.S. Patent Application Publication No. 2012/0193423; U.S. Patent Application Publication No. 2012/0194692; U.S. Patent Application Publication No. 2012/0203647; U.S. Patent Application Publication No. 2012/0223141; U.S. Patent Application Publication No. 2012/0228382; U.S. Patent Application Publication No. 2012/0248188; U.S. Patent Application Publication No. 2013/0043312; U.S. Patent Application Publication No. 2013/0082104; U.S. Patent Application Publication No. 2013/0175341; U.S. Patent Application Publication No. 2013/0175343; U.S. Patent Application Publication No. 2013/0257744; U.S. Patent Application Publication No. 2013/0257759; U.S. Patent Application Publication No. 2013/0270346; U.S. Patent Application Publication No. 2013/0292475; U.S. Patent Application Publication No. 2013/0292477; U.S. Patent Application Publication No. 2013/0293539; U.S. Patent Application Publication No. 2013/0293540; U.S. Patent Application Publication No. 2013/0306728; U.S. Patent Application Publication No. 2013/0306731; U.S. Patent Application Publication No. 2013/0307964; U.S. Patent Application Publication No. 2013/0308625; U.S. Patent Application Publication No. 2013/0313324; U.S. Patent Application Publication No. 2013/0332996; U.S. Patent Application Publication No. 2014/0001267; U.S. Patent Application Publication No. 2014/0025584; U.S. Patent Application Publication No. 2014/0034734; U.S. Patent Application Publication No. 2014/0036848; U.S. Patent Application Publication No. 2014/0039693; U.S. Patent Application Publication No. 2014/0049120; U.S. Patent Application Publication No. 2014/0049635; U.S. Patent Application Publication No. 2014/0061306; U.S. Patent Application Publication No. 2014/0063289; U.S. Patent Application Publication No. 2014/0066136; U.S. Patent Application Publication No. 2014/0067692; U.S. Patent Application Publication No. 2014/0070005; U.S. Patent Application Publication No. 2014/0071840; U.S. Patent Application Publication No. 2014/0074746; U.S. Patent Application Publication No. 2014/0076974; U.S. Patent Application Publication No. 2014/0097249; U.S. Patent Application Publication No. 2014/0098792; U.S. Patent Application Publication No. 2014/0100813; U.S. Patent Application Publication No. 2014/0103115; U.S. Patent Application Publication No. 2014/0104413; U.S. Patent Application Publication No. 2014/0104414; U.S. Patent Application Publication No. 2014/0104416; U.S. Patent Application Publication No. 2014/0106725; U.S. Patent Application Publication No. 2014/0108010; U.S. Patent Application Publication No. 2014/0108402; U.S. Patent Application Publication No. 2014/0110485; U.S. Patent Application Publication No. 2014/0125853; U.S. Patent Application Publication No. 2014/0125999; U.S. Patent Application Publication No. 2014/0129378; U.S. Patent Application Publication No. 2014/0131443; U.S. Patent Application Publication No. 2014/0133379; U.S. Patent Application Publication No. 2014/0136208; U.S. Patent Application Publication No. 2014/0140585; U.S. Patent Application Publication No. 2014/0152882; U.S. Patent Application Publication No. 2014/0158770; U.S. Patent Application Publication No. 2014/0159869; U.S. Patent Application Publication No. 2014/0166759; U.S. Patent Application Publication No. 2014/0168787; U.S. Patent Application Publication No. 2014/0175165; U.S. Patent Application Publication No. 2014/0191684; U.S. Patent Application Publication No. 2014/0191913; U.S. Patent Application Publication No. 2014/0197304; U.S. Patent Application Publication No. 2014/0214631; U.S. Patent Application Publication No. 2014/0217166; U.S. Patent Application Publication No. 2014/0231500; U.S. Patent Application Publication No. 2014/0247315; U.S. Patent Application Publication No. 2014/0263493; U.S. Patent Application Publication No. 2014/0263645; U.S. Patent Application Publication No. 2014/0270196; U.S. Patent Application Publication No. 2014/0270229; U.S. Patent Application Publication No. 2014/0278387; U.S. Patent Application Publication No. 2014/0288933; U.S. Patent Application Publication No. 2014/0297058; U.S. Patent Application Publication No. 2014/0299665; U.S. Patent Application Publication No. 2014/0332590; U.S. Patent Application Publication No. 2014/0351317; U.S. Patent Application Publication No. 2014/0362184; U.S. Patent Application Publication No. 2014/0363015; U.S. Patent Application Publication No. 2014/0369511; U.S. Patent Application Publication No. 2014/0374483; U.S. Patent Application Publication No. 2014/0374485; U.S. Patent Application Publication No. 2015/0001301; U.S. Patent Application Publication No. 2015/0001304; U.S. Patent Application Publication No. 2015/0009338; U.S. Patent Application Publication No. 2015/0014416; U.S. Patent Application Publication No. 2015/0021397; U.S. Patent Application Publication No. 2015/0028104; U.S. Patent Application Publication No. 2015/0029002; U.S. Patent Application Publication No. 2015/0032709; U.S. Patent Application Publication No. 2015/0039309; U.S. Patent Application Publication No. 2015/0039878; U.S. Patent Application Publication No. 2015/0040378; U.S. Patent Application Publication No. 2015/0049347; U.S. Patent Application Publication No. 2015/0051992; U.S. Patent Application Publication No. 2015/0053769; U.S. Patent Application Publication No. 2015/0062366; U.S. Patent Application Publication No. 2015/0063215; U.S. Patent Application Publication No. 2015/0088522; U.S. Patent Application Publication No. 2015/0096872; U.S. Patent Application Publication No. 2015/0100196; U.S. Patent Application Publication No. 2015/0102109; U.S. Patent Application Publication No. 2015/0115035; U.S. Patent Application Publication No. 2015/0127791; U.S. Patent Application Publication No. 2015/0128116; U.S. Patent Application Publication No. 2015/0133047; U.S. Patent Application Publication No. 2015/0134470; U.S. Patent Application Publication No. 2015/0136851; U.S. Patent Application Publication No. 2015/0142492; U.S. Patent Application Publication No. 2015/0144692; U.S. Patent Application Publication No. 2015/0144698; U.S. Patent Application Publication No. 2015/0149946; U.S. Patent Application Publication No. 2015/0161429; U.S. Patent Application Publication No. 2015/0178523; U.S. Patent Application Publication No. 2015/0178537; U.S. Patent Application Publication No. 2015/0178685; U.S. Patent Application Publication No. 2015/0181109; U.S. Patent Application Publication No. 2015/0199957; U.S. Patent Application Publication No. 2015/0210199; U.S. Patent Application Publication No. 2015/0212565; U.S. Patent Application Publication No. 2015/0213647; U.S. Patent Application Publication No. 2015/0220753; U.S. Patent Application Publication No. 2015/0220901; U.S. Patent Application Publication No. 2015/0227189; U.S. Patent Application Publication No. 2015/0236984; U.S. Patent Application Publication No. 2015/0239348; U.S. Patent Application Publication No. 2015/0242658; U.S. Patent Application Publication No. 2015/0248572; U.S. Patent Application Publication No. 2015/0254485; U.S. Patent Application Publication No. 2015/0261643; U.S. Patent Application Publication No. 2015/0264624; U.S. Patent Application Publication No. 2015/0268971; U.S. Patent Application Publication No. 2015/0269402; U.S. Patent Application Publication No. 2015/0288689; U.S. Patent Application Publication No. 2015/0288896; U.S. Patent Application Publication No. 2015/0310243; U.S. Patent Application Publication No. 2015/0310244; U.S. Patent Application Publication No. 2015/0310389; U.S. Patent Application Publication No. 2015/0312780; U.S. Patent Application Publication No. 2015/0327012; U.S. Patent Application Publication No. 2016/0014251; U.S. Patent Application Publication No. 2016/0025697; U.S. Patent Application Publication No. 2016/0026838; U.S. Patent Application Publication No. 2016/0026839; U.S. Patent Application Publication No. 2016/0040982; U.S. Patent Application Publication No. 2016/0042241; U.S. Patent Application Publication No. 2016/0057230; U.S. Patent Application Publication No. 2016/0062473; U.S. Patent Application Publication No. 2016/0070944; U.S. Patent Application Publication No. 2016/0092805; U.S. Patent Application Publication No. 2016/0101936; U.S. Patent Application Publication No. 2016/0104019; U.S. Patent Application Publication No. 2016/0104274; U.S. Patent Application Publication No. 2016/0109219; U.S. Patent Application Publication No. 2016/0109220; U.S. Patent Application Publication No. 2016/0109224; U.S. Patent Application Publication No. 2016/0112631; U.S. Patent Application Publication No. 2016/0112643; U.S. Patent Application Publication No. 2016/0117627; U.S. Patent Application Publication No. 2016/0124516; U.S. Patent Application Publication No. 2016/0125217; U.S. Patent Application Publication No. 2016/0125342; U.S. Patent Application Publication No. 2016/0125873; U.S. Patent Application Publication No. 2016/0133253; U.S. Patent Application Publication No. 2016/0171597; U.S. Patent Application Publication No. 2016/0171666; U.S. Patent Application Publication No. 2016/0171720; U.S. Patent Application Publication No. 2016/0171775; U.S. Patent Application Publication No. 2016/0171777; U.S. Patent Application Publication No. 2016/0174674; U.S. Patent Application Publication No. 2016/0178479; U.S. Patent Application Publication No. 2016/0178685; U.S. Patent Application Publication No. 2016/0178707; U.S. Patent Application Publication No. 2016/0179132; U.S. Patent Application Publication No. 2016/0179143; U.S. Patent Application Publication No. 2016/0179368; U.S. Patent Application Publication No. 2016/0179378; U.S. Patent Application Publication No. 2016/0180130; U.S. Patent Application Publication No. 2016/0180133; U.S. Patent Application Publication No. 2016/0180136; U.S. Patent Application Publication No. 2016/0180594; U.S. Patent Application Publication No. 2016/0180663; U.S. Patent Application Publication No. 2016/0180678; U.S. Patent Application Publication No. 2016/0180713; U.S. Patent Application Publication No. 2016/0185136; U.S. Patent Application Publication No. 2016/0185291; U.S. Patent Application Publication No. 2016/0186926; U.S. Patent Application Publication No. 2016/0188861; U.S. Patent Application Publication No. 2016/0188939; U.S. Patent Application Publication No. 2016/0188940; U.S. Patent Application Publication No. 2016/0188941; U.S. Patent Application Publication No. 2016/0188942; U.S. Patent Application Publication No. 2016/0188943; U.S. Patent Application Publication No. 2016/0188944; U.S. Patent Application Publication No. 2016/0189076; U.S. Patent Application Publication No. 2016/0189087; U.S. Patent Application Publication No. 2016/0189088; U.S. Patent Application Publication No. 2016/0189092; U.S. Patent Application Publication No. 2016/0189284; U.S. Patent Application Publication No. 2016/0189288; U.S. Patent Application Publication No. 2016/0189366; U.S. Patent Application Publication No. 2016/0189443; U.S. Patent Application Publication No. 2016/0189447; U.S. Patent Application Publication No. 2016/0189489; U.S. Patent Application Publication No. 2016/0192051; U.S. Patent Application Publication No. 2016/0202951; U.S. Patent Application Publication No. 2016/0202958; U.S. Patent Application Publication No. 2016/0202959; U.S. Patent Application Publication No. 2016/0203021; U.S. Patent Application Publication No. 2016/0203429; U.S. Patent Application Publication No. 2016/0203797; U.S. Patent Application Publication No. 2016/0203820; U.S. Patent Application Publication No. 2016/0204623; U.S. Patent Application Publication No. 2016/0204636; U.S. Patent Application Publication No. 2016/0204638; U.S. Patent Application Publication No. 2016/0227912; U.S. Patent Application Publication No. 2016/0232891; U.S. Patent Application Publication No. 2016/0292477; U.S. Patent Application Publication No. 2016/0294779; U.S. Patent Application Publication No. 2016/0306769; U.S. Patent Application Publication No. 2016/0314276; U.S. Patent Application Publication No. 2016/0314294; U.S. Patent Application Publication No. 2016/0316190; U.S. Patent Application Publication No. 2016/0323310; U.S. Patent Application Publication No. 2016/0325677; U.S. Patent Application Publication No. 2016/0327614; U.S. Patent Application Publication No. 2016/0327930; U.S. Patent Application Publication No. 2016/0328762; U.S. Patent Application Publication No. 2016/0330218; U.S. Patent Application Publication No. 2016/0343163; U.S. Patent Application Publication No. 2016/0343176; U.S. Patent Application Publication No. 2016/0364914; U.S. Patent Application Publication No. 2016/0370220; U.S. Patent Application Publication No. 2016/0372282; U.S. Patent Application Publication No. 2016/0373847; U.S. Patent Application Publication No. 2016/0377414; U.S. Patent Application Publication No. 2016/0377417; U.S. Patent Application Publication No. 2017/0010141; U.S. Patent Application Publication No. 2017/0010328; U.S. Patent Application Publication No. 2017/0010780; U.S. Patent Application Publication No. 2017/0016714; U.S. Patent Application Publication No. 2017/0018094; U.S. Patent Application Publication No. 2017/0046603; U.S. Patent Application Publication No. 2017/0047864; U.S. Patent Application Publication No. 2017/0053146; U.S. Patent Application Publication No. 2017/0053147; U.S. Patent Application Publication No. 2017/0053647; U.S. Patent Application Publication No. 2017/0055606; U.S. Patent Application Publication No. 2017/0060316; U.S. Patent Application Publication No. 2017/0061961; U.S. Patent Application Publication No. 2017/0064634; U.S. Patent Application Publication No. 2017/0083730; U.S. Patent Application Publication No. 2017/0091502; U.S. Patent Application Publication No. 2017/0091706; U.S. Patent Application Publication No. 2017/0091741; U.S. Patent Application Publication No. 2017/0091904; U.S. Patent Application Publication No. 2017/0092908; U.S. Patent Application Publication No. 2017/0094238; U.S. Patent Application Publication No. 2017/0098947; U.S. Patent Application Publication No. 2017/0100949; U.S. Patent Application Publication No. 2017/0108838; U.S. Patent Application Publication No. 2017/0108895; U.S. Patent Application Publication No. 2017/0118355; U.S. Patent Application Publication No. 2017/0123598; U.S. Patent Application Publication No. 2017/0124369; U.S. Patent Application Publication No. 2017/0124396; U.S. Patent Application Publication No. 2017/0124687; U.S. Patent Application Publication No. 2017/0126873; U.S. Patent Application Publication No. 2017/0126904; U.S. Patent Application Publication No. 2017/0139012; U.S. Patent Application Publication No. 2017/0140329; U.S. Patent Application Publication No. 2017/0140731; U.S. Patent Application Publication No. 2017/0147847; U.S. Patent Application Publication No. 2017/0150124; U.S. Patent Application Publication No. 2017/0169198; U.S. Patent Application Publication No. 2017/0171035; U.S. Patent Application Publication No. 2017/0171703; U.S. Patent Application Publication No. 2017/0171803; U.S. Patent Application Publication No. 2017/0180359; U.S. Patent Application Publication No. 2017/0180577; U.S. Patent Application Publication No. 2017/0181299; U.S. Patent Application Publication No. 2017/0190192; U.S. Patent Application Publication No. 2017/0193432; U.S. Patent Application Publication No. 2017/0193461; U.S. Patent Application Publication No. 2017/0193727; U.S. Patent Application Publication No. 2017/0199266; U.S. Patent Application Publication No. 2017/0200108; and U.S. Patent Application Publication No. 2017/0200275.

In the specification and/or figures, typical embodiments of the present invention have been disclosed. The present invention is not limited to such exemplary embodiments. The use of the term “and/or” includes any and all combinations of one or more of the associated listed items. The figures are schematic representations and so are not necessarily drawn to scale. Unless otherwise noted, specific terms have been used in a generic and descriptive sense and not for purposes of limitation.

Claims

1. A method for Optical Character Recognition (OCR), the method comprising: obtaining a plurality of OCR decoding results, wherein each OCR decoding result comprises a plurality of positions;calculating, for each of the plurality of OCR decoding results, a number of occurrences of a first character in a first position of the plurality of positions and the number of occurrences of a second character in the first position;determining that the first character in the first position of a first OCR decoding result of the plurality of OCR decoding results is not in a predefined lexicon and incrementing the number of occurrences of the first character;associating the first character with an individual confidence score based on the number of occurrences of the first character;associating the second character with the individual confidence score based on the number of occurrences of the second character; anddetermining a presumptively valid OCR decoding character result for the first position based on the individual confidence score.

2. The method according to claim 1, wherein obtaining the plurality of OCR decoding results further comprises: capturing a plurality of images of the same one or more OCR characters; anddecoding each of the plurality of images to a character in each of the plurality of positions.

3. The method according to claim 1, further comprising: calculating, for each of the plurality of OCR decoding results, the number of occurrences of each additional character in each remaining position of the plurality of positions, wherein the number of occurrences of each additional character is incremented in an instance in which a character is not in the predefined lexicon;associating each additional character in each remaining position with the individual confidence score; anddetermining the presumptively valid OCR decoding character result for each remaining position based on the individual confidence score.

4. The method according to claim 2, further comprising: associating each presumptively valid OCR decoding character result with a total confidence score based on the individual confidence score; andselecting a final OCR decoding result that is associated with a highest total confidence score.

5. The method according to claim 2, further comprising: selecting an identified character for each position of the OCR decoding result based on the identified character in each position having a highest total confidence score for that position across each OCR decoding character result.

6. The method according to claim 1, further comprising: determining that the number of occurrences of the first character or the number of occurrences of the second character do not satisfy a particular threshold; andobtaining additional OCR decoding results.

7. The method according to claim 1, further comprising: determining that the number of occurrences of the first character and the number of occurrences of the second character do not satisfy a particular threshold; andobtaining additional OCR decoding results.

8. The method according to claim 1, further comprising: determining that the individual confidence score of the first character does not satisfy a minimum individual confidence score; andexcluding the first character.

9. An optical character recognition (OCR) apparatus comprising: an image sensor for capturing one or more images; a memory; anda processor communicatively coupled to the image sensor to convert the one or more images into a plurality of OCR decoding results; anda non-transitory memory including computer program instructions configured to, when executed by the processor, cause the OCR apparatus to: calculate, for each of the plurality of OCR decoding results, a number of occurrences of a first character in a first position of a plurality of positions and the number of occurrences of a second character in the first position;determine that the first character in the first position of a first OCR decoding result of the plurality of OCR decoding results is not in a predefined lexicon and incrementing the number of occurrences of the first character;associate the first character with an individual confidence score based on the number of occurrences of the first character;associate the second character with the individual confidence score based on the number of occurrences of the second character; anddetermine a presumptively valid OCR decoding character result for the first position based on the individual confidence score.

10. The OCR apparatus according to claim 9, wherein the non-transitory memory including the computer program instructions is further configured to, when executed by the processor, cause the OCR apparatus to: calculate, for each OCR decoding result of the plurality of OCR decoding results, the number of occurrences of each additional character in each remaining position of the plurality of positions, wherein the number of occurrences of each additional character is incremented in an instance in which a character is not in the predefined lexicon;associate each additional character in each remaining position with the individual confidence score; anddetermine the presumptively valid OCR decoding character result for each remaining position based on the individual confidence score.

11. The OCR apparatus according to claim 10, wherein the non-transitory memory including the computer program instructions is further configured to, when executed by the processor, cause the OCR apparatus to: associate each presumptively valid OCR decoding character result with a total confidence score based on the individual confidence score; andselect a final OCR decoding result that is associated with a highest total confidence score.

12. The OCR apparatus according to claim 10, wherein the non-transitory memory including the computer program instructions is further configured to, when executed by the processor, cause the OCR apparatus to: select an identified character for each position of the OCR decoding result based on the identified character in each position having a highest total confidence score for that position across each OCR decoding character result.

13. The OCR apparatus according to claim 9, wherein the non-transitory memory including the computer program instructions is further configured to, when executed by the processor, cause the OCR apparatus to: determine that the number of occurrences of the first character or the number of occurrences of the second character do not satisfy a particular threshold; andobtain additional OCR decoding results.

14. The OCR apparatus according to claim 9, wherein the non-transitory memory including the computer program instructions is further configured to, when executed by the processor, cause the OCR apparatus to: determine that the number of occurrences of the first character and the number of occurrences of the second character do not satisfy a particular threshold; andobtain additional OCR decoding results.

15. The OCR apparatus according to claim 9, wherein the non-transitory memory including the computer program instructions is further configured to, when executed by the processor, cause the OCR apparatus to: determine that the individual confidence score of the first character does not satisfy a minimum individual confidence score; andexclude the first character.

16. A non-transitory computer program product for Optical Character Recognition (OCR), wherein the non-transitory computer program product comprises: a non-transitory computer readable storage medium having computer readable program code embodied in said non-transitory computer readable storage medium, said computer readable program code comprising steps of: obtaining a plurality of OCR decoding results, wherein each OCR decoding result comprises a plurality of positions;calculating, for each of the plurality of OCR decoding results, a number of occurrences of a first character in a first position of the plurality of positions and the number of occurrences of a second character in the first position;determining that the first character in the first position of a first OCR decoding result of the plurality of OCR decoding results is not in a predefined lexicon and incrementing the number of occurrences of the first character;associating the first character with an individual confidence score based on the number of occurrences of the first character;associating the second character with the individual confidence score based on the number of occurrences of the second character; anddetermining a presumptively valid OCR decoding character result for the first position based on the individual confidence score.

17. The non-transitory computer program product according to claim 16, wherein said computer readable program code further comprising the steps of: capturing a plurality of images of the same one or more OCR characters; anddecoding each of the plurality of images to a character in each of the plurality of positions.

18. The non-transitory computer program product according to claim 16, wherein said computer readable program code further comprising the steps of: calculating, for each of the plurality of OCR decoding results, the number of occurrences of each additional character in each remaining position of the plurality of positions, wherein the number of occurrences of each additional character is incremented in an instance in which a character is not in the predefined lexicon;associating each additional character in each remaining position with the individual confidence score; anddetermining the presumptively valid OCR decoding character result for each remaining position based on the individual confidence score.

19. The non-transitory computer program product according to claim 18, wherein said computer readable program code further comprising the steps of: associating each presumptively valid OCR decoding character result with a total confidence score based on the individual confidence score; andselecting a final OCR decoding result that is associated with a highest total confidence score.

20. The non-transitory computer program product according to claim 18, wherein said computer readable program code further comprising a step of: selecting an identified character for each position of the OCR decoding result based on the identified character in each position having a highest total confidence score for that position across each OCR decoding character result.