JPS60201484A - Picture data processing unit - Google Patents

Abstract

(57) [Summary] This bulletin contains application data before electronic filing, so abstract data is not recorded.

Description

DETAILED DESCRIPTION OF THE INVENTION (a) Technical Field of the Invention The present invention relates to a data processing apparatus, and more particularly to an image data processing apparatus that performs processing for emphasizing the edges of an image.

(b) Background of the technology In the field of image processing technology that uses an autopsy system, images to be processed are observed and image information such as density is converted into digital data for each pixel arranged in a mesh shape with a predetermined pinch. After conversion, various arithmetic processing is performed, and the multivalued image data is made clearer so that it is easier for the viewer to see. Alternatively, processing for extracting the outline of an image is often performed for the purpose of feature extraction in image recognition.

However, the image to be processed may have gradual changes in shading from the beginning, and even if the image has steep changes in shading, a dull observed image may be obtained if the observation system lacks high-frequency characteristics. It is often difficult to quantify or extract contours.

Therefore, if you create an image with gradual changes in shading like this,
Value. Alternatively, when extracting the contour of such an image, a process called edge enhancement is performed in order to clarify the contour of one image, that is, the change in density.

(C) Prior Art and Problems Conventionally, image data processing apparatuses for performing the edge enhancement described above use linear filter circuits such as differential filter circuits, or outputs from various types of filter circuits corresponding to edge directions. A non-linear filter circuit that selects one of these has been used.

However, in the former case, if there is isolated noise as shown in Figure 1 (al) in the image to be processed, edge enhancement produces an output as shown in Figure 1 (bl), so a high-quality edge enhancement effect can be obtained. The disadvantage is that it cannot be done.

Furthermore, the latter method has disadvantages in that it requires processing time and increases the scale of the device since it uses a wide variety of filter circuits, for example, eight types.

(d) Object of the Invention An object of the present invention is to provide an image data processing device that can obtain high-quality edge enhancement effects in a relatively short processing time.

(e) Structure of the invention That is, the image data processing device according to the present invention is as follows.

The value of the read pixel obtained by reading the image data obtained by digitizing the image pixel by pixel arranged in a predetermined pinch mesh shape one by one is calculated as the maximum of the values of multiple pixels within a predetermined range including that pixel. a maximum value filter circuit that converts the value of the read pixel to a value and outputs the value; a minimum value filter circuit that converts the value of the read pixel to the minimum value of the values of a plurality of pixels within a predetermined range including the pixel and outputs the value; and the read pixel a differential filter circuit that performs differential operation on the value of by pixel values within a predetermined range including the pixel and outputs the result, an output of the maximum value filter circuit, an output of the minimum value filter circuit, and an output of the differential filter circuit. The above object is achieved by including an arithmetic circuit that performs an arithmetic operation on the value of the read pixel using the above.

(f) Examples of the invention Next, the gist of the present invention will be specifically explained using examples.

Fig. 2 is a block diagram showing the configuration of an embodiment of the present invention. In Fig. 9, 1 observes the image to be processed and converts the density of each pixel arranged in a mesh pattern with a pitch of 0.1 mm into multivalued data. An image memory 2 stores multivalued image data obtained by conversion, and 2 is a 3×
A differential filter circuit 3 outputs a second-order differential value by performing neighborhood operations using 1 and 1×3 masks, 3 is constituted by a shift register, and reads the multivalued image data stored in the image memory 1 by mask scanning. Maximum value filter circuit 4 and minimum value filter circuit 5 described later
A time delay circuit 4 outputs a time delay corresponding to the processing time in , and 4 reads the multivalued image data stored in the image memory 1 by mask scanning and creates a 3×1 or i×3 mask centered on each pixel. A maximum value filter circuit 5 outputs the maximum value of pixel values in pixels in a neighboring area of
A minimum value filter circuit that outputs the minimum value of the pixel values in the pixels in the vicinity of the ×3 mask.

6 sequentially inputs the second-order differential value obtained by the differential filter circuit 2, and determines whether the value is larger than th with respect to a preset threshold th, or is a value between th and -th, or A determination circuit 7 determines whether the value is smaller than -th, and 7 selects the output of the maximum value filter circuit 4 when the output of the differential filter circuit 2 is a value greater than th, according to the determination result by the determination circuit 6. When the output of the differential filter circuit 2 is between th and -th, the output of the time delay circuit 3 is selected and output, and the output of the differential filter circuit 2 is smaller than -th. 8 is an output memory that does not store the output of the arithmetic circuit 7.

Note that the mask used in the maximum value filter circuit 4 and the minimum value filter circuit 5 is that the differential filter circuit 2 is
When a mask of 1 is used, 3×1 is used, and when a mask of 2×3 differential filter circuits is used, 1×3 is used. Further, each mask is used by rask scanning in which the longitudinal direction is the main scanning direction.

For ease of understanding, Figure 4 is a diagram that explains the principle of operation in each part using image data expressed in a -dimensional manner (the horizontal axis indicates the pixel reading direction, and the vertical axis indicates the density). (a) is an example of a processed image with a relatively steep gradient of density before being observed to be stored in the image memory 1, and (b) is an example of a processed image with a relatively steep gradient of density (al). (C) is the image data obtained by differentiating the image of (1)) with the differential filter circuit 2 and then i, (d) is image data obtained by processing the image of (bl by the minimum value filter circuit 5), (e) is image data obtained by processing the image of fbl by the 'li 4 in the maximum value filter circuit.

(fl is the image data obtained by delaying the image of [bl by the time delay circuit 3 and fg. Also, fgl shows the output of the arithmetic circuit 7 that appears as the image of fb).

In Fig. 4 fgl, ■, ■ and ■ indicate that the output of the one-time delay circuit 3 (dl/output fe of the maximum value filter circuit 4) and the output ffl of the minimum value filter circuit 5 are selected in the arithmetic circuit 7, respectively. represents that.

By comparing fbl and fffl, it can be seen that an image with a gradual a change (bl) is an image with a steep density change (gl conversion).

In the above embodiment, since the output of the differential filter circuit 2 does not appear as it is as the output of the arithmetic circuit 7, there is no adverse effect caused by isolated noise as in the conventional example explained with reference to FIG.

Note that in the above embodiment, the differential filter circuit 2, maximum value filter circuit 4, and minimum value filter circuit 5 all perform neighborhood calculations using 3×1 and 1×3 windows. Similar processing can also be performed using neighborhood calculations using various windows.

(g) Effects of the Invention As explained in the examples, according to the present invention, a high-quality edge enhancement effect can be obtained in a relatively short processing time, and contour extraction of multi-level images can be performed accurately. can.

[Brief explanation of drawings]

FIG. 1 is an explanatory diagram regarding the problem of the conventional example, FIG. 2 is a block diagram of an embodiment of the present invention, FIG. 3 is an explanatory diagram of a differential filter circuit, and FIG. 4 is an explanatory diagram regarding the operation of the embodiment. In the figure, 2 is a differential filter circuit, 4 is a maximum value filter circuit, 5 is a minimum value filter circuit, and 7 is an arithmetic circuit. #I @ thing 2 closed

Claims (2)

[Claims] (1) Digitize each pixel of the image that can be arranged in a predetermined mesh shape and read the image data by 1 pixel, and calculate the value of the read pixel for multiple values within a predetermined range including that pixel. A maximum value filter circuit that converts the value of the read pixel to the maximum value and outputs it, and a minimum value filter circuit that converts the value of the read pixel to the minimum value of the values of a plurality of pixels within a predetermined range including that pixel and outputs it. and a differential filter circuit that performs a differential operation on the value of the read pixel using values of pixels within a predetermined range including the read pixel and outputs the result. An image data processing device comprising: an arithmetic circuit that performs an arithmetic operation on the value of the read pixel using the output of the maximum value filter circuit, the output of the minimum value filter circuit, and the output of the differential filter circuit. . (2) The arithmetic circuit performs a process of selecting either the value of the read pixel, the output of the maximum value filter circuit, or the output of the minimum value filter circuit based on the output of the differential filter circuit. An image data processing device according to claim (1), characterized in that: