JPH0362178A - Picture processor - Google Patents

Abstract

PURPOSE:To simply obtain a picture processing device which can separate distinctly parts resembling in color by dividing a picture element level value corresponding to an original picture by each picture element level value of a picture obtained by giving smoothing processing and further normalizing processing to the original picture. CONSTITUTION:An eyeground is photographed by the CCD camera 2 of an eyeground camera 1, and after an obtained eyeground picture is digitized, and is stored in a picture memory 5, arithmetic processing is executed. At first, an original eyeground picture A is read into a picture processor 4. Next, a smoothing filter F is set to all the picture elements of the picture A, and the smoothing processing to substitute the mean value of all the picture element levels of a picture element area is executed, and the eyeground picture B is obtained. Next, the maximum value MAX among all the picture element levels of the picture B is selected, and the normalizing processing to divide all the picture element levels of the picture B by the value MAX is executed, and the eyeground picture C in which the value MAX of the picture element level is made '1' is obtained. Next, all the picture element levels of the unprocessed picture A in the memory 5 are divided by the picture element levels at the corresponding position of the picture C, and the eyeground picture D is obtained. When the picture D is monitored, a blood vessel image P can be observed as being separated from an eyeground retina image Q.

Description

DETAILED DESCRIPTION OF THE INVENTION [Industrial Application Field] The present invention relates to an image processing device that processes an original image to make it clear.

[Prior Art] Conventional 1 For example, in order to improve the contrast of the fundus image, the density gradation of the fundus image is changed using a lookup table, or in order to correct an image with biased density gradation. The histogram flattening process used is being performed.

[Problem to be solved by the invention] In addition to the original image, a background image, that is, an image showing brightness bias over a wide range, is taken for a general image, and the brightness bias is corrected over a wide range. However, it is not possible to capture a background image of the fundus of the eye to be examined.Therefore, it is conceivable to perform the processing in the conventional example described above on the fundus image.

However, in the fundus image processed in the conventional example described above, it is impossible to perform density emphasis change processing or histogram flattening processing using a lookup table to correct partial brightness bias, so one color is It is difficult to clearly separate fundus TSM images and blood vessel images, which are similar.

The object of the present invention is the above-mentioned lWl! It is an object of the present invention to provide an image processing device that can resolve the aa problem and clearly separate, for example, parts with similar colors through easy processing.

[Means for Solving the Problems] In order to achieve the above object, an image processing device according to the present invention includes a first calculation means for smoothing an original image, and processing by the first calculation means. a second calculation means for normalizing the image processed by the second calculation means; and a third calculation means for dividing the level value of each pixel of the original image by the level value at the same position of the image processed by the second calculation means. It is characterized by having a calculation means.

[Operation] The image processing device having the above configuration calculates the corresponding pixel of the original image based on each pixel level value of the image, which is obtained by performing smoothing processing and normalization processing on the original image that has been digitally input and stored, for example. Divide the level value.

C Example] The present invention will be explained in detail based on illustrated embodiments.

FIG. 1 shows a configuration diagram when the image processing device according to the embodiment of the present invention is applied to an ophthalmological measuring device.
A CD camera 2 is provided, an image processing processor 4 is connected to an A/D converter 3 connected to the CCD camera 2, and an image memory 5 is connected to the image processing processor 4.
and an image monitor 6 are connected.

Then, the fundus is photographed by the CCD camera 2 of the fundus camera 1, and the obtained fundus image is digitized via the A/D converter 3 and stored in the image memory 5, and then the operations shown in the flowchart of FIG. In the flowchart of FIG. 2 for performing starvation, first, in step 11, the original fundus image A, which has already been stored in the image memory 5, is read into the image processing processor 4.Next, in step 12, the original fundus image A is read into the image processing processor 4. As shown in Fig. 3, a smoothing filter F with a size of, for example, 15 x 15 pixels is applied to all pixels of A fundus image B is obtained by performing smoothing processing to replace the pixel area with the average value of all pixel levels. Thereby, a background image using the original fundus image A can be obtained.

Next, in step 13, the maximum value WAX among the levels of all pixels of this average image B is selected, and in step 14, a normalization process is performed in which the levels of all pixels of fundus image B are divided by the maximum value WAX, A fundus image C is obtained using the maximum pixel level MAX. Furthermore, in step 15, all pixel levels of the original fundus image A, which is stored in the image memory 5 and has not been processed, are divided by the level of pixels at the corresponding position of the fundus image C to obtain a fundus image. By performing this division process, it is guaranteed that the maximum level of the fundus image A is equal to or lower than the maximum level of the original fundus image A.

In the next step 16, the fundus image is displayed on the image monitor 6.
If the blood vessel image P is output as shown in FIG. 4, the blood vessel image P is clearly separated from the fundus retinal image Q and can be easily observed.

If a smoothing filter with a size larger than the 15×15 pixel smoothing filter F used in the above embodiment is used, effective smoothing processing can be performed in a shorter time. Even with this smoothing filter, the blood vessel image can be clearly separated from the fundus image without any practical problems, and if the smoothing process is repeated several times in history, a more preferable image can be obtained.

Further, the original fundus image A to be processed does not have to be an image photographed by the CCD camera 2 of the fundus camera 1 as in the above embodiment, but may be a fundus image digitally input to the image memory 5 from a slide scanner, or A/A / fundus images recorded on analog media such as still videos and video recorders
The fundus image may be digitally input to the image memory 5 after being D-converted.

[Effects of the Invention] As explained above, the image processing device according to the present invention performs division processing of an original image by an image that has been subjected to smoothing processing in which a smoothing filter is applied to the original image and normalization processing. Therefore, shading is corrected and the entire image can be binarized using a certain threshold value, and for example, a retinal image and a blood vessel image in a fundus image can be clearly distinguished.

[Brief explanation of drawings]

The drawings show an embodiment of the image processing apparatus according to the present invention, in which Fig. 1 is a configuration diagram of the apparatus, Fig. 2 is a flowchart of the processing, Fig. 3 is an explanatory diagram of the smoothing filter, and Fig. 41ffl shows the image after processing. FIG. 2 is an explanatory diagram of a fundus image of FIG. Symbol l is a fundus camera, 2 is a CCD camera, 3 is an A/D converter, 4 is an image processing processor, 5 is an image memory,
6 is an image monitor.

Claims (1)

[Claims] 1. A first calculation means for smoothing the original image, a second calculation means for normalizing the image processed by the first calculation means, and each pixel of the original image. The level value of
an ophthalmological image processing apparatus, comprising: a third calculation means for performing division processing by a level value at the same position of the image processed by the calculation means; 2. The image processing apparatus according to claim 1, wherein the smoothing process by the first calculation means is performed multiple times.