Methods or arrangements for reading or recognising printed or written characters or for recognising patterns

GPHYSICS G06Computing G06KRECOGNITION OF DATA PRESENTATION OF DATA RECORD CARRIERS HANDLING RECORD CARRIERS

G06K9/00Methods or arrangements for reading or recognising printed or written characters or for recognising patterns

G06K9/00006Acquiring or recognising fingerprints or palmprints G06K9/00013Image acquisition G06K9/0002by non-optical methods G06K9/00026by combining adjacent partial images ( G06K9/00033by direct reading G06K9/0004by using electro-optical elements or layers G06K9/00046by using geometrical optics G06K9/00053Protecting the fingerprint sensor against damage caused by the finger G06K9/00067Preprocessing; Feature extraction (minutiae) G06K9/00073Extracting features related to minutiae and pores G06K9/0008Extracting features related to ridge properties; determining the fingerprint type G06K9/00087Matching; Classification G06K9/00093Matching features related to minutiae and pores G06K9/001Matching features related to ridge properties or fingerprint texture G06K9/00107Detecting the live character of the finger G06K9/00114using image processing G06K9/0012using acquisition arrangements G06K9/00127Acquiring and recognising microscopic objects G06K9/00134Acquisition G06K9/0014Pre-processing G06K9/00147Matching; Classification G06K9/00154Reading or verifying signatures; Writer recognition G06K9/00161based only on signature image G06K9/00167based only on signature signals such as velocity or pressure G06K9/00174Acquisition G06K9/00181Matching; classification G06K9/00187Preprocessing; feature extraction G06K9/00194Sampling; contour coding; stroke extraction G06K9/00201Recognising three-dimensional objects G06K9/00208by matching two-dimensional images to three-dimensional objects G06K9/00214by matching three-dimensional models G06K9/00221Acquiring or recognising human faces, facial parts, facial sketches, facial expressions G06K9/00228Detection; Localisation; Normalisation G06K9/00234using pixel segmentation or colour matching G06K9/00241using holistic features G06K9/00248using facial parts and geometric relationships G06K9/00255using acquisition arrangements G06K9/00261using comparisons between temporally consecutive images G06K9/00268Feature extraction; Face representation G06K9/00275Holistic features and representations G06K9/00281Local features and components; Facial parts ; Occluding parts G06K9/00288Classification G06K9/00295of unknown faces, i.e. recognising the same non-enrolled faces G06K9/00302Facial expression recognition G06K9/00308Static expression G06K9/00315Dynamic expression G06K9/00335Recognising movements or behaviour G06K9/00342Recognition of whole body movements G06K9/00348Recognition of walking or running movements G06K9/00355Recognition of hand or arm movements G06K9/00362Recognising human body or animal bodies G06K9/00369Recognition of whole body G06K9/00375Recognition of hand or arm G06K9/00382Static hand biometric or posture recognition G06K9/00389Static hand gesture recognition G06K9/00402Recognising digital ink G06K9/00409Preprocessing; Feature extraction G06K9/00416Sampling; contour coding; stroke extraction G06K9/00422Matching; classification G06K9/00429using a special pattern or subpattern alphabet G06K9/00436using human interaction G06K9/00442Document analysis and understanding; Document recognition G06K9/00449Layout structured with printed lines or input boxes G06K9/00456Classification of image contents G06K9/00463Document analysis by extracting the geometrical structure G06K9/00469Document understanding by extracting the logical structure G06K9/00476Reading or recognising technical drawings or geographical maps G06K9/00483Document matching G06K9/00496Recognising patterns in signals and combinations thereof G06K9/00503Preprocessing G06K9/0051Denoising G06K9/00516by applying a scale-space analysis G06K9/00523Feature extraction G06K9/0053by analysing the shape of a waveform G06K9/00536Classification; Matching G06K9/00543by matching peak patterns G06K9/0055by matching signal segments G06K9/00557by plotting the signal segments against each other G06K9/00563by applying autoregressive analysis G06K9/0057Source localisation; Inverse modelling G06K9/00577Recognising objects characterised by unique random properties, i.e. objects having a physically unclonable function [PUF] G06K9/00597Acquiring or recognising eyes G06K9/00604Acquisition G06K9/0061Preprocessing; Feature extraction G06K9/00617Matching; Classification G06K9/00624Recognising scenes G06K9/0063Recognising patterns in remote scenes G06K9/00637of urban or other man made structures G06K9/00651of network patterns, such as roads, rivers G06K9/00657of vegetation G06K9/00664Recognising scenes such as could be captured by a camera operated by a pedestrian or robot, including objects at substantially different ranges from the camera G06K9/00671for providing information about objects in the scene to a user G06K9/00677Analysis of image collections based on shared content G06K9/00684Categorising the entire scene G06K9/00691Indoor scenes G06K9/00697Outdoor scenes G06K9/00704Urban scenes G06K9/00711Recognising video content G06K9/00718Higher-level, semantic clustering, classification or understanding of video scenes G06K9/00724of sport video content G06K9/00731of news video content G06K9/00744Extracting features from the video content G06K9/00751Detecting suitable features for summarising video content G06K9/00758Matching video sequences G06K9/00765Segmenting video sequences G06K9/00771Recognising scenes under surveillance G06K9/00778Recognition or static of dynamic crowd images G06K9/00785Recognising traffic patterns acquired by static cameras G06K9/00791Recognising scenes perceived from the perspective of a land vehicle G06K9/00798Recognition of lanes or road borders G06K9/00805Detecting potential obstacles G06K9/00812Recognition of available parking space G06K9/00818Recognising traffic signs G06K9/00825Recognition of vehicle or traffic lights G06K9/00832Recognising scenes inside a vehicle G06K9/00838Recognising seat occupancy G06K9/00845Recognising the driver's state or behaviour G06K9/00852Recognising whole cursive words G06K9/00859using word shape G06K9/00865using stroke segmentation G06K9/00872with lexical matching G06K9/00879with probabilistic networks G06K9/00885Biometric patterns not provided for under G06K9/00006, G06K9/00154, G06K9/00335, G06K9/00362, G06K9/00597; Biometric specific functions not specific to the kind of biometric G06K9/00892Use of multiple biometrics G06K9/00899Spoof detection G06K9/00906Detection of body part being alive G06K9/00912Interactive means for assisting the user in correctly positioning the object of interest G06K9/00919Static means for assisting the user in correctly positioning the object of interest G06K9/00926Maintenance of references; Enrolment G06K9/00973Hardware and software architectures for pattern recognition G06K9/00979structured as a network G06K9/00986using specific electronic processors G06K9/00993Management of recognition tasks G06K9/03Detection or correction of errors G06K9/033with the intervention of an operator G06K9/036Evaluation of quality of acquired pattern G06K9/18using printed characters having additional code marks or containing code marks G06K9/183Characters composed of bars G06K9/186Recognition of characters printed with magnetic ink G06K9/20Image acquisition G06K9/2009Construction of image pick-up using regular bi-dimensional dissection G06K9/2018Identifying/ignoring parts by sensing at different wavelengths G06K9/2027Illumination control G06K9/2036Special illumination such as grating, reflections, deflections G06K9/2054Selective acquisition/locating/processing of specific regions G06K9/2063based on a marking or identifier characterising the document or the area G06K9/2072based on positionally close symbols G06K9/2081based on user interaction G06K9/209Sensor details G06K9/22using hand-held instruments G06K9/222the instrument generating sequences of position coordinates corresponding to handwriting; preprocessing or recognising digital ink G06K9/224in three dimensions G06K9/228Hand-held scanners; Optical wands G06K9/24Construction of the instrument G06K9/26using a slot moved over the image G06K9/28using discrete sensing elements at predetermined points G06K9/30using automatic curve following means G06K9/32Aligning or centering of the image pick-up or image-field G06K9/3208Orientation detection or correction G06K9/3216by locating a pattern G06K9/3233Determination of region of interest G06K9/3241Recognising objects as potential recognition candidates based on visual cues G06K9/325Detection of text region in scene imagery, real life image or Web pages G06K9/3258Scene text G06K9/3266Overlay text G06K9/3275Inclination (skew) detection or correction of characters or of image to be recognised G06K9/3283of characters or characters lines G06K9/34Segmentation of touching or overlapping patterns in the image field G06K9/342Cutting or merging image elements G06K9/344using recognition of characters or words G06K9/346Removing patterns interfering with the pattern to be recognised, such as ruled lines, underlines G06K9/348using character size, text spacings, pitch estimation G06K9/36Image preprocessing G06K9/38Quantising the analogue image signal G06K9/40Noise filtering G06K9/42Normalisation of the pattern dimensions G06K9/44Smoothing or thinning of the pattern G06K9/46Extraction of features or characteristics of the image G06K9/4604Detecting partial patterns G06K9/4609by matching or filtering G06K9/4614filtering with Haar-like subimages G06K9/4619Biologically-inspired filters G06K9/4623with interaction between the responses of different filters G06K9/4628Integrating the filters into a hierarchical structure G06K9/4633by mapping characteristic values of the pattern into a parameter space G06K9/4638by analysing connectivity relationships of elements of the pattern G06K9/4642by performing operations within image blocks or by using histograms G06K9/4647summing image-intensity values; Projection and histogram analysis G06K9/4652related to colour G06K9/4661related to illumination properties G06K9/4671Extracting features based on salient regional features G06K9/4676Extracting features based on a plurality of salient regional features G06K9/468related to a structural representation of the pattern G06K9/4685Syntactic representation G06K9/469Graphical representation G06K9/48by coding the contour of the pattern contour related features or features from contour like patterns G06K9/481using vector-coding G06K9/482analysing the spectrum of the contour G06K9/50by analysing segments intersecting the pattern segments obtained by the intersection of the pattern with a scanning pattern G06K9/52by deriving mathematical or geometrical properties from the whole image G06K9/522Frequency domain transformation; Autocorrelation G06K9/525Computation of moments G06K9/527Scale-space domain transformation G06K9/54Combinations of preprocessing functions G06K9/56using a local operator G06K9/58using optical means G06K9/60Combination of image acquisition and preprocessing functions G06K9/605the combination being locally implemented within the sensing arrangement G06K9/62Methods or arrangements for recognition using electronic means G06K9/6201Matching; Proximity measures G06K9/6202Comparing pixel values or logical combinations thereof, or feature values having positional relevance G06K9/6203Shifting or otherwise transforming the patterns to accommodate for positional errors G06K9/6204Matching of contours G06K9/6205by mapping curve parameters onto an accumulator array G06K9/6206involving a deformation of the sample or reference pattern; Elastic matching G06K9/6207based on a local optimisation criterion, e.g. "snakes" G06K9/6209based on shape statistics G06K9/621based also on statistics of image patches G06K9/6211Matching configurations of points or features G06K9/6212Comparing statistics of pixel or of feature values G06K9/6214based on a parametric eigenspace representation G06K9/6215Proximity measures G06K9/6217Design or setup of recognition systems and techniques; Extraction of features in feature space; Clustering techniques; Blind source separation G06K9/6218Clustering techniques G06K9/6219Hierarchical techniques G06K9/622Non-hierarchical partitioning techniques G06K9/6221based on statistics G06K9/6222with an adaptive number of clusters G06K9/6223with a fixed number of clusters G06K9/6224based on graph theory G06K9/6226based on the modelling of probability density functions G06K9/6227Selection of pattern recognition techniques G06K9/6228Selecting the most significant subset of features G06K9/6229by using evolutionary computational techniques G06K9/623by ranking or filtering the set of features G06K9/6231by evaluating different subsets according to an optimisation criteria such as class separability G06K9/6232Extracting features by transforming the feature space G06K9/6234based on a discrimination criterion G06K9/6235Rendering the within-class scatter matrix nonsingular G06K9/6239based on a naturality criterion G06K9/624based on a separation criterion G06K9/6242of statistical independence G06K9/6243of decorrelation or non-stationarity G06K9/6244enforcing sparsity or involving a domain transformation G06K9/6245characterised by a domain transformation G06K9/6246overcoming non-stationarity or permutations G06K9/6247based on an approximation criterion G06K9/6248nonlinearly G06K9/6249based on a sparsity criterion G06K9/6251based on a criterion of topology preservation G06K9/6252involving differential geometry G06K9/6253User interactive design ; Environments; Tool boxes G06K9/6254Interactive pattern learning with a human teacher G06K9/6255Determining representative reference patterns G06K9/6256Obtaining sets of training patterns; Bootstrap methods G06K9/6257characterised by the organisation or the structure of the process G06K9/6259characterised by the incorporation of unlabelled data G06K9/626Selecting classification rules G06K9/6261partitioning the feature space G06K9/6262Validation, performance evaluation or active pattern learning techniques G06K9/6263based on the feedback of a supervisor G06K9/6264the supervisor being an automated "intelligent" module G06K9/6265based on a specific statistical test G06K9/6267Classification techniques G06K9/6268relating to the classification paradigm G06K9/6269based on the distance between the decision surface and training patterns lying on the boundary of the class cluster G06K9/627based on distances between the pattern to be recognised and training or reference patterns G06K9/6271based on distances to prototypes G06K9/6272based on distances to cluster centroïds G06K9/6273Smoothing the distance G06K9/6274based on distances to neighbourhood prototypes G06K9/6276based on distances to closest patterns G06K9/6277based on a parametric (probabilistic) model G06K9/6278Bayesian classification G06K9/6279relating to the number of classes G06K9/628Multiple classes G06K9/6281Piecewise classification G06K9/6282Tree-organised sequential classifiers G06K9/6284Single class perspective G06K9/6285relating to the decision surface G06K9/6286Linear G06K9/6287Non-linear G06K9/6288Fusion techniques, i.e. combining data from various sources G06K9/6289of input or preprocessed data G06K9/629of extracted features G06K9/6292of classification results G06K9/6293of classification results relating to different input data G06K9/6296Graphical models G06K9/6297Markov models and related models G06K9/6298Statistical pre-processing G06K9/64using simultaneous comparisons or correlations of the image signals with a plurality of references G06K9/645using a resistor matrix G06K9/66references adjustable by an adaptive method G06K9/68using sequential comparisons of the image signals with a plurality of references in which the sequence of the image signals or the references is relevant G06K9/6807Dividing the references in groups prior to recognition, the recognition taking place in steps; Selecting relevant dictionaries G06K9/6814according to the graphical properties G06K9/6821Alphabet recognition G06K9/6828Font recognition G06K9/6835Discrimination between machine-print, hand-print and cursive writing G06K9/6842according to the linguistic properties G06K9/685Involving plural approaches G06K9/6857Coarse/fine approaches G06K9/6878Syntactic or structural pattern recognition G06K9/6885Syntactic analysis G06K9/6892Graph matching G06K9/70the selection of the next reference depending on the result of the preceding comparison contains no documents G06K9/72using context analysis based on the provisionally recognized identity of a number of successive patterns G06K9/723Lexical context G06K9/726Syntactic or semantic context G06K9/74Arrangements for recognition using optical reference masks G06K9/741using frequency domain filters G06K9/743characterised by the kind of filter G06K9/745the filter being related to phase processing G06K9/746the filter being related to the combination of filters G06K9/748using spatial domain filters G06K9/76using holographic masks G06K9/78Combination of image acquisition and recognition functions G06K9/80Combination of image preprocessing and recognition functions G06K9/82using optical means in one or both functions