Patent Application
20070071417
The present invention relates to a method for the transfer of a data stream of full digital images generated by a digital image detection unit with a first predetermined number of picture elements via an interface to an image recording unit. The invention is furthermore directed to a corresponding digital image detection unit.
The HD (high definition) motion picture cameras widespread today are only insufficiently suitable to be able to generate digital images with an aspect ratio of 2.35:1 (Cinemascope process) for film production. In accordance with the HD standard, these cameras record an image format of 16×9 so that, when spherical lenses are used, digital images are generated with an image format of 16×9, which corresponds to an image aspect ratio of 1.778:1.
These images can only be converted into the Cinemascope format in that substantial horizontal image portions are cut off at the upper and lower side of the image. However, this is associated with an unwanted reduction in the vertical resolution of the resulting image.
Whereas typically so-called anamorphic lenses with a squeeze of 2:1 are used in the Cinemascope process to squeeze a scene with an image ratio of 2.35:1 onto a 4×3 sensor (or onto a 35 mm film in analog processes), so-called semi-anamorphic optical systems, which do not correspond to the conventional standard and are therefore currently not available, would have to be used for the imaging of a corresponding scene on an HD sensor with an image format of 16×9. In addition to the costs required for the development of such lenses, images taken using such a semi-anamorphic lens would create a different impression than images taken with conventional anamorphic lenses so that the acceptance of such semi-anamorphic lenses is questionable.
Since the taking of images in the 4×3 format, optionally with anamorphic lenses for the generation of the Cinemascope format has been widespread both with analog and, in the meantime, with digital taking processes, for a long time, the further use of digital motion picture cameras with a 4×3 sensor and conventional anamorphic lens elements for the taking of films in the Cinemascope process is desirable.
Since, in the meantime, the HD format with an image format of 16×9 has become established for the storage and the playing back of digitally encoded films, the normally available standardized image recording units, for example in the form of hard disk recorders or tape recording devices, are also usually only suitable for the recording of digital images in the 16×9 format.
Starting from an identical number of horizontal picture elements of, for example, 1920 pixels per image both for the 4×3 format and for the 16×9 format, the problem thus results of how the larger data volume of a digital image present in the 4×3 format can be recorded and stored by a recording device working according to the HD standard. The problem furthermore occurs that, in accordance with the HD SDI standard (high definition serial digital interface standard), data transfer rates are only possible with which no full images taken in the 4×3 format can be transmitted in real time (>25 frames per second).
The same problems exist when images of a film typically present in 35 mm (4×3 format) scanned using a film scanner should be recorded using a conventional HD recorder without any quality loss, i.e. without any reduced image information.
It is the object of the present invention to provide a method and an apparatus with which it is possible to transmit full digital images, in particular in the 4×3 format, generated with a first number of picture elements to an image recording device and to store them in it at full resolution and without a reduction in image quality, with the image recording unit only being made for the storage of digital images with a second number of picture elements reduced with respect to the first number of picture elements, in particular as a conventional HD compatible recording device.
In accordance with the invention, this object is satisfied by a method for the transfer of a data stream of full digital images generated by a digital image detection unit with a first predetermined number of picture elements via an interface to an image recording unit, with the interface being made for the transfer of at least one data stream of digital images with a second number of picture elements different from the first number of picture elements and the image recording unit being made for the storage of digital images with the second number of picture elements and with the second number of picture elements being lower than the first number of picture elements, wherein the full digital images generated with the first number of picture elements are in each case divided into two part digital images, with the part images each having a number of picture elements which is used for the storage of image information and which is lower than or equal to the second number of picture elements, and with the image information stored in the two part images together containing the total information content of the original full image, two data streams of the digital part images being generated, with in each case a part image of a full digital image being associated with the first data stream and the other part image of the full digital image being associated with the second data stream, the two data streams of the part digital images being transmitted to the image recording unit via the interface and the transmitted part digital images being stored with the second number of picture elements separately in each case in the image recording unit.
A digital image detection unit in accordance with the invention comprises a sensor for the generation of full digital images with a first predetermined number of picture elements, an interface for the transfer of at least one data stream of digital images with a second number of picture elements different from the first number of picture elements to an image recording unit, with the second number of picture elements being lower than the first number of picture elements, an image division unit which is made for the division of the full images generated into two respective part digital images, with the part images each having a number of picture elements used for the storage of image information which is lower than or equal to the second number of picture elements and with the image information stored in the two part images together containing at least the total information content of the original full image, a data stream generation unit for the generation of two data streams of the part digital images, with one respective part image of a full digital image being associated with the first data stream and the other part image of the full digital image being associated with the second data stream, and a control unit for the control of the transfer of the two data streams of the part digital images to the image recording unit via the interface.
In accordance with the invention, the recorded images generated with the first number of picture elements are thus divided into two part digital images whose number of picture elements actually used for the storage of image information is lower than or equal to the second number of picture elements, i.e. that number of picture elements with which the images can be stored in the image recording unit at least along the vertical or the horizontal direction. In this connection, all the picture elements of the two part images therefore do not have to be used for the storage of image information from the full image. It is, however, important that the digital full images are divided into the two part digital images such that the total information content of the original full image is maintained in the two part images so that a later reconstruction of the original full image is possible without any loss of quality and with the original resolution. Since usual HD compatible recording units often have two HD SDI compatible inputs for the parallel recording of two HD SDI compatible data streams, such standardized recording units can be used without change. It is generally also possible to use two recording devices separate from one another of which each records one of the two data streams of the part digital images. It is necessary in this context that the two recording devices or the respective data streams are synchronized with one another so that a precise time association of the part images generated is ensured.
The recording of full digital images in the 4×3 format on standardized HD compatible recording units is thus possible with the invention without any loss in quality and reduction in the resolution.
A particular advantage of the invention consists of the fact that the said division of the full images (recorded images) into the two respective part images can take place such that one of the two part images can be used as a useful image for a check of the image information of the total image and the other part image can be used as an additional image for the provision of the supplementary image information. In other words, the named useful image can be observed without further processing on a 16×9 monitor, for example, in order to hereby make a first inspection of the recorded images.
The number of picture elements is to be understood in connection with the invention as the number of the picture elements in the vertical direction (=number of lines), or the number of the picture elements in the horizontal direction (=number of columns) or the total number of the picture elements of the image in question.
In accordance with an advantageous embodiment of the invention, the number of picture elements of the part images is matched to the second number of picture elements—i.e. the number of picture elements of the image recording unit—on the division of the data stream into the part images when the number of picture elements of the part images actually used for the storage of image information is lower than the second number of picture elements. If the part images generated on the division of the data stream are already present with the second number of picture elements, a corresponding matching is no longer necessary since the part images can in this case be transferred unchanged to the image recording unit working with the second number of picture elements and can be stored or recorded in it. A matching to the second number of picture elements can take place, for example, in that “black bars” are generated at the upper and lower margins and/or at the left hand and right hand margins of the digital images.
In accordance with a further preferred embodiment of the invention, the part digital images generated from a full digital image have the same number of picture elements. A simple reconstruction of the part digital images stored in the image recording unit is thereby possible.
The full digital images with the first number of picture elements are preferably generated in a standardized 4×3 format. Depending on the number of pixels of the sensor used, the first number of picture elements can amount, for example, to 1920×1440 pixels or, for example, to 2880×2160 pixels.
The part digital images are advantageously stored in a standardized 16×9 format in the image recording unit. If the part digital images are already present in the 16×9 format after their generation, the part digital images can thus already be transferred in a standardized 16×9 format to the image recording unit via the interface. A typical resolution for a standardized 16×9 format is the HD format with 1920×1080 pixels.
In accordance with a further advantageous embodiment of the invention, the full digital images with the first number of picture elements have a predetermined number of lines and a predetermined number of columns, with in each case all even-numbered lines of the full digital image being associated with a part image of the first data stream and all the odd-numbered lines of the full digital image being associated with a part image of the second data stream on the division of the full digital images.
This special division of the full digital images into two respective part digital images has the advantage that, on the recording of the full images using an anamorphic optical system, the part image appears as a non-distorted 2.35:1 image within a 16×9 image (so-called “letter box”) when viewed on a 16×9 monitor.
In a further preferred embodiment of the invention, the full digital images with the first number of picture elements have a predetermined number of lines and a predetermined number of columns, with in each case a predetermined number of lines from the middle region of the full digital image being associated with a part image of the first data stream and the remaining, upper and lower lines of the full digital image being associated with a part image of the second data stream on the division of the full digital images into the part images. The number of lines corresponding to the second number of picture elements can preferably be used as the predetermined number of lines from the middle region. It is furthermore advantageous for the middle region of the full digital image to be selected to be symmetrical to the horizontal central axis of the full digital image.
On this division of the full digital image into part digital images, when a spherical lens is used, the middle region of the taken image is stored in the respective first part image such that the stored first part images are suitable alone for reproduction with the second resolution, for example in the 16×9 format. Since no material objects are usually contained in the upper and lower regions of the taken image, a digital image detection unit in accordance with the invention can in this case be used as a “normal” HD motion picture camera.
The possibility exists, when an anamorphic lens is used with the same taking unit (camera), to make an expansion of the image to the Cinemascope format by means of the upper and lower edge bars stored in the second part images.
The remaining upper and lower lines of the full digital image are preferably associated with the part image of the second data stream such that they also form the upper and/or the lower lines of the second part digital image. The unused pixels lying between the upper and lower lines of the second part digital image can be masked or marked with “black”.
In accordance with a further advantageous embodiment of the invention, the full digital images with the first number of picture elements have a predetermined number of lines and a predetermined number of columns, in each case with a predetermined number of lines from the upper region of the full digital image being associated with a part image of the first data stream and the remaining, lower lines of the full digital image being associated with a part image of the second data stream on the division of the full digital images into the part images.
Depending on the application, a spherical lens or an anamorphic lens, in particular with a distortion factor of 2:1, is advantageously used for the generation of the full digital images.
The digital image detection unit can be made, for example, as a digital motion picture camera or also as a film scanner. The interface is advantageously made in accordance with the HD SDI standard to permit an unproblematic connection to the recording units compatible with the HD SDI standard.
The invention will be described in more detail in the following with reference to embodiments and to the drawings, in which are shown:
FIGS. 2 to 4 three further embodiments for the division of a full digital image into two part digital images;
When an anamorphic lens with the distortion factor 2:1 is used, the image to be taken in the Cinemascope format of 2.35:1 is imaged into a full digital image with 1692×1440 picture elements so that respectively unused perpendicular bars remain with a width of 114 picture elements at the right hand and at the left hand margin of the sensor surface. The corresponding picture elements are either not read out or can be marked with “black”, for example, so that two perpendicular black bars arise in the full digital image 1.
In accordance with the invention, the correspondingly taken full digital images 1 are divided into two part digital images 3, 4 as is indicated by arrows 5, 6.
The division takes place in accordance with
With a line number of a total of 1440 lines of the full digital image 1, 720 image lines are thus associated with each of the two part images 3, 4, as can be seen from
Since the part images with a height of 720 image lines do not completely fill the HD image format with a height of 1080 image lines, horizontally extending black bars with a height of in each case 180 image lines are present at the upper and at the lower margins of the part images 3, 4.
The part images 3, 4 generated in this manner thus have an HD conforming resolution of 1920×1080 and can be transferred in two separate data streams together with respectively following part images 3, 4 via an HD-SDI conforming standard interface to an HD compatible recording unit and can be stored separately there in the standard HD format.
As can be recognized from
It is generally possible with this embodiment as also with all further embodiments for the total width of 1920 pixels of the taking sensor to be used, which results in an image aspect ratio of 2.667:1 when an anamorphic lens with a distortion factor of 2:1 is used. In this case, the part images of the two data streams each have an effective resolution of 1920×720 pixels (picture elements actually used for the storage of image information). The standardized resolution of 1692×720 picture elements or of the recombined squeezed image of 1692×1440 picture elements can then be established in the post-production.
In the example of
In accordance with
It is advantageous with this division that the first part digital image 11 contains the material part of the full digital image 10 in the standardized 16×9 format and the part images 11 can therefore be viewed without any further processing on a 16×9 monitor. The full digital images 10 can in turn be produced without any loss in quality and be shown in the 4×3 format after corresponding recombination of the part images 11 with the respective part images 12 due to the two part digital images 12.
The same also applies to the division in accordance with
The invention does not only prove to be advantageous when—as shown in FIGS. 1 to 4—the total number of picture elements of the part images 3, 4, 11, 12, 14, 15, 19, 20 is lower than the total number of picture elements of the respective full image 1, 10, 13, or 18. The explained advantage of the maintenance of quality is also achieved when the number of picture elements of the part images is only lower along the vertical or the horizontal direction than with the full image, as can occur, for example, on a conversion from a 4×3 format to a 16×9 format.
In
The sensor 23 is connected at its output to an image division unit 24 by which the full digital images generated by the sensor 23 are divided into two part digital images and are subsequently output for the generation of a data stream so that the image division unit 24 simultaneously forms a data stream generation unit. The two data streams of part digital images generated in this manner are transmitted to two outputs 25, 26, in particular HD SDI compatible outputs, of an interface 27, with the transfer of the two data streams of the part digital images being controlled via a control unit 28.
The two digital data streams are transferred via data cables 29, 30 to HD SDI conforming inputs 31, 32 of an HD recorder 33 which is capable of the simultaneous recording of two parallel data streams of part digital images, in particular in the 16×9 format.
The HD recorder can record the part digital images, for example, on a hard disk 34 or on any other suitable medium such as a magnetic tape or an optical disk.
Alternatively, it is shown in the right hand region in
It is shown in
| Number | Date | Country | Kind |
|---|---|---|---|
| 10 2005 037 282.1 | Aug 2005 | DE | national |
