NL9301907A - Method for electronically storing radiation images - Google Patents

Abstract

Method for electronically storing radiation images, in which a radiation image is projected onto a radiation- sensitive sensor which optionally carries a pattern of radiation-sensitive cells distributed over its surface. Said sensor emits an analog or digital signal which is representative for said radiation image, said signal, if it is an analog signal, being digitized. Care is taken to ensure that the digital signal or the signal obtained by means of digitization, optionally with an intervening signal translation step, is representative of an electronic image composed of alternating positive and negative parallel lines which are to be projected so as to be substantially equally spaced and are each divided into a multiplicity of positions whose radiation value has been determined separately. Said signal for said electronic image is processed to produce a storage signal which is stored in a memory, said storage signal being determined by means of the following steps in succession.

Description

Title: Method for electronically storing radiation images.

The invention relates to the electronic storage of images for, for example, documentation, archiving and later use. Such images are, for example, photographic images of both visible light and, for example, X-rays.

It is common to convert images into an electronic signal for display with a picture tube of, for example, a conventional television set. For example, a camera with a photosensitive sensor is used to take a picture of an object or image. The analog signal obtained is again visible with a picture tube. To store the signal (possibly temporarily), so-called magnetic tapes can be used, which are suitable for the storage of analog signals. However, a magnetic tape has a sequential effect, so that retrieving a certain image is difficult and time consuming.

Attempts have been made to store the signals in an easier-to-access memory medium, so that each image is separately directly addressable, and thus can be retrieved more quickly. In principle, the use of so-called memory chips or memory disks lend themselves to this, which can be used in conjunction with a computing processor, for instance a so-called personal computer. For this purpose, a digital or digitized signal must be used. It has been found, however, that an image recorded with, for example, a conventional video camera, preview with a conventional TV set with, for example, black and white PAL or NTSC, or SECAM system, for direct, i.e. unedited, storage, particularly large memory space of more than 100 kB (yte), in particular 400 kfi, is required. As a result, with the currently available, direct-addressable, digital storage media, for which the cost price increases exponentially with increasing storage capacity, storing large amounts of images is extremely expensive.

Attempts have been made to limit the required storage space for an image. The present invention aims to make a contribution to television.

To this end, the invention proposes to process the signals before saving according to claim 1. Preferably, this is based on a recording suitable for displaying on a picture tube, for which said recording is shown in parallel, alternating positive and negative lines over the height of the picture, wherein each line is divided into so-called pixels, each of which can be controlled separately according to brightness and possibly color. Assuming the PAL standard sample tubes, there are 288 positive and negative lines each, with 720 pixels per line.

It has been found that, in particular in the case of X-ray images of dentures read in with a video camera, with the omission of lines and pixels as indicated in claim 1, a signal results which guarantees sufficient image sharpness and details when displayed via a picture tube. However, this storage signal has significantly less data compared to the original signal, in particular to less than at least 1/9 of the original amount of data. The required memory space is also substantially limited accordingly.

According to a further advantageous embodiment of the invention, the memory space required for the storage of the storage signal is further limited by coding for brightness. According to an advantageous variant, the gray value difference with the adjacent pixel is determined for further limiting the required memory space of all pixels of the processed signal. The frequency of each gray value difference is then determined. The greyscale difference with the lowest frequency gets the longest code. The encoding table is also saved.

According to a further development of the invention, a standard coding table for the gray value differences is determined in order to limit the required computing time. To this end, the frequency distribution of the gray values is determined for a large number of corresponding images, for example orthodontic X-rays. On the basis thereof, an average frequency distribution is determined, which is used as a reference for drawing up a standard coding table. Then grayscale images, which are also housed in the image group, are encoded with that standard encoding table. For all images, therefore, only one coding table needs to be stored, which can further reduce the required memory space. In addition, such a standard encoding table can also be stored in another medium, for example hardware, or in a (C.D.) ROM memory.

Both of the aforementioned compression steps using the coding of the gray level differences between adjacent pixels, the image signal is reduced from 405 kB initially, such as

common with a black and white image in the PAL system, up to about 20kB. It the current 3 * 5 "diskettes for personal computers (eg the IBM

A.T.), with a storage capacity of 1 MB or more, as a result, more than 50 images can be stored on one diskette, sufficient for the full administration of, for example, a dentist.

The storage signals obtained according to the invention are stored in an electronic memory, for example in a memory chip or a diskette. Storage on a memory tape (tape) is of course also possible.

Thus, the invention provides an efficient process for processing image signals into relatively few memory capacity-consuming storage signals. The use of particularly simple means is thereby required. The signal processing and processing can be done with a simple personal computer, connected to a conventional video camera for standard TV. applications.

If the invention is used, for example, for the storage of dental X-rays, an incidentally light-tight holder or adapter can be attached to the video camera in register with the lens and in the exposure path thereof, in which an X-ray photo can be placed opposite the lens. The adapter is transparent to the camera lens. For example, the adapter is a tubular or sleeve-shaped body of revolution, which is connected to the camera with a head end, and a photo can be attached on the opposite end side with fasteners located on the adapter (for example insertion slots running parallel to the photo edges for receiving therein). those photo edges The adapter can be adjusted so that the X-ray is always at a fixed distance from the camera lens in or on the adapter The distance can also be adjustable, for example, with a lengthwise bellow-shaped spacer on the adapter. thereby adjusted so that the X-ray image is projected full-screen on the photosensitive pick-up element of the camera, which is exposed with a light source on the side facing away from the lens of the X-ray image (X-rays are photo-translucent as photo negatives). from at v. gas discharge lamps, such as the well-known fluorescent lamps or mercury vapor lamps. The adapter may also have a window fitted in a suitable place for illuminating the X-ray with ambient light (daylight or working light). As a result, a relatively small investment of video camera, adapter and PC is required. required, with the video camera and PC. can also be used for other purposes due to its multifunctional character. Furthermore, the described arrangement is simple and efficient, and does not require any special knowledge for its construction and operation.

Some tests have been carried out on the following test set-up:

An X-ray image was recorded full screen with a Sony TR705SE video camera, suitable for the PAL system and with its lens in macro mode. The lens used has a focal length f = 7.8 mm. Distance from photo to front of camera body was adjustable. The X-ray material was of the fast type, as is customary in dental practices today.

For full-frame reading of a photo I with a size I of 61 * 46 mm, the distance from the photo to the front of the camera housing was 50 mm. With the PAL system, the photo was thus scanned in horizontal direction at 11.8 pixels per mm. photo width. In the vertical direction, the scan was 12.5 lines per mm. photo height.

A next photo 11 with size 11 of 80 * 60 mm was also read in full screen under otherwise the same conditions, with the exception of the distance to the camera lens. Obviously, compared to size I, the scan was less pixels resp. line per mm photo length.

A further photo III with size III of 46 * 35 mm was also read in full screen under otherwise the same conditions, with the exception of the distance to the camera lens. Obviously, in comparison with size I, the scan was with more pixels resp. lines permm photo length.

In all cases, the size of the direct read-in signal was approximately 405 KB. In accordance with the invention, the size of the signal was always reduced to a storage signal of about 20kB. This storage signal was always displayed on the same monitor, maintaining all other settings during the test. The reproduction of the storage signal of photo I turned out to be of good quality, with sufficient distinguishable details. The same reproduction of photo II turned out to be of lesser quality; that of better quality photo III. These comparative tests showed that reading in photos with less than 7 pixels and lines per mm. photo length yielded a qualitatively unsatisfactory image of the storage signal obtained therefrom with the invention on a monitor.

It is also possible to read the image in a so-called image scanning plate, in which, possibly through a lens, in the manner as with a copier or a facsimile device, the object to be imaged is projected on a CCD chip with the same pixel pattern as that of the screen. The best results are obtained when the object is as full-screen as possible.

Of course, images other than from X-rays can also be processed with the invention. For example, color or black-and-white photos, but also "live ** images electronically recorded directly with the camera. Direct reading of eg X-rays with an X-ray sensitive CCD chip, as already applied in the dental world, can be used with the invention. Digitization can take place with the photosensitive sensor.

Claims (8)

A method of electronically storing radiation images, wherein a radiation image is projected onto a radiation sensitive sensor, optionally with a pattern of radiation sensitive cells spread over the surface, which sensor outputs an analog or digital signal representative of that radiation image, which signal, when this is an analog signal, is digitized, and care is taken that the digital signal or the signal obtained with digitizing, possibly through a signal translation step, is representative of an electronically constructed image of alternating positive and negative parallel, substantially equidistant spacing lines to be projected from one another, each of which is divided into a plurality of positions whose radiation value is determined separately, which signal for that electronic image is processed into a storage signal which is stored in a memory, characterized in that said storage signal 1 is determined by successive steps: A. all positive or negative lines are omitted; B. of all remaining lines (positive or negative), the number of positions is substantially halved by omitting every other position; C. of all remaining lines, every third line is omitted; D. of each line, the number of positions is substantially reduced to 2/3 by omitting one of the two positions. A method according to claim 1, wherein the number of positive and negative lines is equal and each is 250 to 300, in particular 280 to 290, more in particular 288, and the number of positions on each line is equal and 65 to 800, in particular especially 700 to 750, more in particular 720. A method according to any one of the preceding claims, wherein the following further steps are carried out: E. the radiation value difference between the radiation values thereof is determined from two adjacent positions in one line; F. the data in the storage signal about the position pairs referred to under E. is replaced by that for the data radiation value difference. The method according to claim 3, wherein the further subsequent steps are carried out: G. the frequency of all radiation value differences is determined for all radiation value differences; H. the radiation frequency differences with the highest frequency are encoded with the shortest storage code. Method according to claim 3 or 4, in which a storage signal is determined from a predetermined number of similar radiation images using steps A to F, after which the following steps are taken: I. the frequency of the radiation value differences is determined for each storage signal; K. of all storage signals an average frequency spread for radiative value differences is determined, which frequency spread is stored as a reference; and L. the radiation value differences for each subsequent storage signal to be determined are encoded using that reference frequency distribution. A method according to any preceding claim, wherein the sensor is sensitive to shades of gray. A method according to any preceding claim, wherein the radiation images projected onto the sensor are from X-rays, such as orthodontic X-rays. A method according to any one of the preceding claims, wherein a two-dimensional image exposed to visible light, such as an X-ray, is projected on a conventional video camera for use with a conventional TV set with a CCD chip sensor and an intermediate image and that sensor has a lens, wherein that lens and the distance between the image and the lens are set so that the image is completely full screen. Method according to one of the preceding claims, wherein the radiation image is imaged on the radiation-sensitive sensor such that it is 1 mm horizontally. length of the image with at least 7. in particular 9. more particularly reads 11 pixels on one horizontal line, and in vertical direction 1 mm. reads the length of the image by at least 7. in particular 9. more in particular 11 vertical lines one above the other, for example in accordance with the PAL system.