JPH0223779A - Picture processing device - Google Patents

Abstract

PURPOSE:To attain real-time spatial filter processing by performing operation necessitated for spatial filter processing at a time severely for each line of a filter pattern. CONSTITUTION:The picture elements of one line portion are written from an original picture respectively in each line memory L1 to Lm. When these are read out, the memories L1 to Lm are started to read at a time, and every picture element is multiplied individually by a corresponding factor by each multiplier. After that, they are added successively by each adder after adjusting timing by each delay element. Further, the sum of each line is added successively by additional adders provided in the memories after the memory L2, and is outputted as a filter image. Thus, since mXn items necessitated for generating the filter image can be obtained at a time for every picture element, the real- time spatial filter processing becomes possible.

Description

[Detailed Description of the Invention] [Object of the Invention] (Industrial Application Field) The present invention relates to an image processing device that creates a display image by performing image reconstruction processing based on a video signal from an imaging device.
In particular, the present invention relates to improvements in spatial filter processing necessary for image sharpening, smoothing, and feature extraction based on data obtained by A/D conversion of video signals.

(Prior Art) Generally, in an image processing device, spatial filter processing is performed to sharpen, smooth, and extract features of an image based on data obtained by A/D converting a video signal. Do it as follows.

Conventional image processing devices are equipped with a frame memory for storing images, a memory controller for shifting images, a screening circuit, etc., and the entire original image stored in the frame memory corresponds to the position of the filter pattern. The filter image was created by temporarily storing each term obtained by multiplying the coefficients of the filter pattern and each pixel in the image memory in the frame memory, and then adding the necessary terms. .

For example, if the original image shown in Figure 3 is
When processing with ×3 filter pattern, each pixel of the filter image is 'A'cll"C,ab"C+3c"Cab"
C,, xt Cah' ↑C3, & ・CAJ,C
3A"4gu'・ C11b life C, Yuot C13dt
C3,E CJ"cBrown'L-"Cal l
"CJA" + C""...1) Therefore, in order to obtain one pixel of the filter image, the filter coefficient and the original image pixel must be multiplied nine times. Calculate this term by term and store it in the image memory, and after all terms have been found, add it 8 times according to formula 1) for the IC that calculates the filter image, and perform this operation for each pixel. I was creating a filter image.

(Problem to be Solved by the Invention) In the conventional image processing device as described above, a huge amount of calculations necessary for creating a filter image by spatial filter processing is sequentially performed, so real-time spatial filter processing is not necessary. It was possible.

This means that sufficient spatial filter processing cannot be performed in the case of continuous moving image display, and therefore, such moving images have no choice but to have low resolution.

The present invention has been made in view of the above circumstances, and an object of the present invention is to provide an image processing device that can perform real-time spatial filter processing.

[Structure of the Invention] (Means for Solving the Problems) In order to solve the above-mentioned problems, the present invention provides storage means having a memory capacity equal to the number of columns for each row of a matrix-like filter pattern, and a storage means having a memory capacity equal to the number of columns thereof, and The image processing apparatus is equipped with a spatial filter processing mechanism equipped with an equal number of calculation means.

(Function) In the image processing apparatus to which the present invention is applied, the spatial filter processing mechanism performs calculations necessary for spatial filter processing for each row of the filter pattern at once, so that real-time spatial filter processing is possible.

(Embodiment) FIG. 1 is a block diagram of an embodiment of an image processing apparatus to which the present invention is applied.

This embodiment includes an A/D converter 1 that digitizes a video signal from an imaging device, a frame memory 2 that stores the digitized video signal, and an ALU (ALU) that performs image reconstruction processing while using the frame memory 2. Arit
hmetic Logic Unit) 3, real-time filter unit 4 that performs spatial filter processing,
An image enhancement circuit 5 that performs image enhancement processing, a D/A converter 6 that converts the image signal into an analog signal and sends it to a display device, and a CPU (Central Pr) that controls the work of each of the above components.
It consists of 7 units.

Note that details of the real-time filter unit 4 will be described later.

In this embodiment having such a configuration, a video signal from an X-ray diagnostic apparatus is digitized by an A/D converter 1 and stored in a frame memory 2. It is assumed that the frame memory 2 also includes an image memory used for temporal storage in the middle of the image reconstruction process. The stored video signal is sent to the ALU
3 and the image memory of this frame memory 2, image reconstruction processing is performed, but spatial filter processing and image enhancement processing are not performed here.

The original image obtained from image reconstruction processing using the frame memory 2 and ALU 3 is subjected to spatial filter processing in real time in a real-time filter unit 4, as described later, and then subjected to image enhancement processing in an image enhancement circuit 5. The converter 6 converts it into an analog signal and sends it to the CRT as an image signal.
(Cathod RaVTube) or the like is sent to a display device. Note that each of these tasks is performed under the control of the CPU 7.

Next, the real-time filter unit 4 will be explained. FIG. 2 shows the configuration of the real-time filter unit 4. Here, it is assumed that the filter pattern of the spatial filtering process is an mxn matrix.

The real-time filter unit 4 is equipped with m line memories 11 to Lll+ corresponding to each row of the filter pattern. These line memories 1 to Ln+ each have a storage capacity for one line, and each pixel read from the original image in the frame memory 2 is stored in the line memory 1 for each line.
1 to lm. The real-time filter unit 4 is also equipped with n multipliers, n adders ((n-1) only in L1), and (n-1) delay elements for each line memory L1 to L1m. , these are arranged as shown in FIG.

Pixels for one line are written to each line memory L1 to Ll11 from the original image, but when reading out, each line memory L1 to lm starts reading at the same time, and each pixel is individually written to each line. After being multiplied by the filter coefficient corresponding to -C by the multiplier, the sum is added sequentially by each adder in synchronization with the delay element, and the total of each row is further stored in the line memory 1.
They are sequentially added by additional adders installed below 2 and output as a filter image. As a result, the mxn terms necessary for creating a filter image can be found for each pixel at once, making real-time spatial filter processing possible.

Note that m and n vary depending on the filter pattern used, but various filter patterns can be used by setting them to the maximum size filter pattern and operating the required number of filter patterns smaller than that. It becomes like this.

As described above, in this embodiment, spatial filter processing is performed in real time by the action of the real time filter unit 4.

[Effects of the Invention] As explained above and described in 1c, in the image processing device to which the present invention is applied, the spatial filter processing mechanism performs the calculations necessary for spatial filter processing for each row of the filter pattern at the same time, thereby performing real-time spatial filtering. processing becomes possible. This makes it possible to obtain resolution similar to that of still images even in moving images.

[Brief explanation of the drawing]

Fig. 1 is a block diagram of an embodiment of the present invention, Fig. 2 is a diagram showing the configuration of a spatial filter processing circuit on the same side, and Fig. 3 explains spatial filter processing used in a conventional image processing device. This is a diagram. 1...A/D converter 2...Frame memory 3...ALU 4...Real time filter unit 5...Image enhancement circuit 6...D/A converter 7...CPU ■11 Konori Right

Claims (1)

[Claims] For each row of a matrix filter pattern used for spatial filter processing for sharpening, smoothing, and feature extraction of an image based on data obtained by A/D converting an image signal,
An image processing apparatus comprising a spatial filter processing mechanism equipped with a storage means having a memory capacity equal to the number of columns and arithmetic means whose quantity is equal to the number of columns.