JPS63249273A - Discriminating system for whether picture memory is good or not - Google Patents

Abstract

PURPOSE:To shorten the discrimination time by binarizing picture data for discrimination after storing this data in a picture memory where memory cells are arranged in a matrix and comparing results from addition of data in respective rows and columns of the matrix with results from the same processing of picture data for discrimination before storage. CONSTITUTION:After picture data for discrimination is stored in the picture memory where memory cells are arranged in a matrix, this data is read out and is binarized to obtain a matrix where each matrix element is '0' or '1'. An XY projection circuit is used to perform the addition processing in each row and column of the matrix, and addition results of respective rows and columns are obtained. Consequently, if the number of areas of the matrix as the discrimination object is mXn, m-number of sums are obtained with respect to row, and n-number of sums are obtained with respect to column, and (m+n)-number of addition results are obtained in total. These results are compared and collated with (m+n)-number of addition results of data, which is obtained by subjecting picture data for discrimination before storage in the picture memory to the same operation processing, in respective rows and columns of the matrix, thereby discriminating whether the picture memory is good or not.

Description

[Detailed Description of the Invention] (Summary) The present invention uses an XY projection circuit to temporarily store image data for pass/fail judgment in an image memory in which memory cells are arranged in a matrix, read it out, and perform binarization processing. By doing so, the result obtained by adding each row and column of the matrix is compared with the result obtained by the same processing on the judgment image data before being stored in the image memory to determine the quality. Therefore, the amount of data to be compared is small, so the judgment time is shortened.
The bus load on the processor device will be reduced.

(Industrial Application Field) The present invention relates to a method for determining the quality of an image memory, and in particular, a predetermined quality determination method for an image memory in which memory cells for storing image data in units of a predetermined word length are arranged in a matrix. The present invention relates to a method for determining the quality of an image memory, which stores image data for use in image memory and determines the quality of the image memory.

(Prior Art) Conventionally, as a method for determining the quality of an image memory, predetermined image data for quality determination is stored once in each memory cell of the image memory in units of a predetermined word length, and then read out by a processor device. There is a method for determining the quality of the image memory by comparing read image data with the image data for determination before being stored in the image memory for each word length unit (pixel) stored in the memory cell. Ta.

[Problem that the invention seeks to solve]

By the way, the image memory is, for example, 2.048x 2.04
Assuming that 8 memory cells are arranged and each memory cell can store data representing the brightness of a pixel for each 8-bit word length unit, it has a large capacity i of about 4 MB. become.

Therefore, comparing the read image data for judgment and the image data for quality judgment before being stored in the image memory via the processor device one by one for each memory cell places a large load on the bus of the processor device. This will require a lot of time to process the pass/fail judgment.

SUMMARY OF THE INVENTION The present invention has been made in order to solve the above-mentioned problems, and an object of the present invention is to provide a method for determining the quality of an image memory, which does not place a large load on the bus of a processor device and has a short determination processing time. It has been accomplished.

(Means for Solving the Problems) In order to solve the above problems, the present invention provides a system in which memory cells for storing image data in units of a predetermined word length are arranged in a matrix as shown in FIG. In the method of storing predetermined quality judgment image data in the image memory (S1) and determining the quality of the image memory, the once stored judgment image data is read out and the data is read out for each memory cell. is subjected to certain arithmetic processing and binarized (S2
), the binarized data is added for each row and column of the matrix using an XY projection circuit (
S3), the quality of the image memory is determined by comparing the addition result with the result obtained by the same process as the process for the determination image data before being stored in the image memory (S4). .

[Effect]

In the present invention, in step S1, image data for quality determination (memory patrol/check data) is transferred to memory cells in an area targeted for quality determination in an image memory in which memory cells are arranged in a matrix every predetermined tone unit. Write to and store. The image data once written is read out in step S2, and is binarized by performing certain arithmetic processing on each image data stored in each memory cell, so that each matrix element is 0 or 1. Get the matrix.

In step S3, addition processing is performed for each row and column of the matrix using the XY projection circuit on the binarized data to obtain addition results for each row and column.

Therefore, when the area of the matrix to be judged is mXn, m row sums are obtained, and n column sums are obtained, so that metric+nn addition results are obtained.

In step S4, the data obtained by performing the same arithmetic processing as the binarization processing described above on the quality judgment image data before being stored in the image memory for each word length unit is added for each row and column of the matrix described above. The quality of the image memory is determined by comparing and checking the m+n results obtained.

(Example) Next, embodiments according to the present invention will be described.

FIG. 2 shows a circuit configuration diagram realizing the method of this embodiment.

The camera 1 is used to obtain image data from a subject, and is used to decompose an image into pixels of each 8-bit word length unit and transfer the image data to an image memory. The image may be one that changes over time, or one that does not change.

The selector 2 selects image data from either a processor device 12 such as a CPU or the camera 1. When determining the quality of the image memory 3, image data for quality determination from the processor device 12 is selected. The image memory 3 has a plurality of memory cells arranged in a matrix, for example 2,048 memory cells, which store image data pixel by pixel.
It is composed of 2,048 memory cells, and each memory cell stores image data representing the brightness of each pixel using 8 bits. That is, the image memory 3 used has a capacity of approximately 4 Mbytes.

The X-address counter 4 indicates the storage position of image data for the rows of memory cells arranged in a matrix, and by counting a predetermined clock signal output from the memory control section 7. It generates an X address. On the other hand, the Y address counter 5 indicates the storage position of the image data for the column of the memory cell, and
The Y address is generated by counting the CARRY signal of the X address as a clock signal that is output every time the address specification is completed.

The access area control unit 6 uses the X address and Y address counter 5 which are the output results of the X address counter 4.
The image memory access area (area) is controlled based on the Y address which is the output result of the image memory. In other words, the access position of the image memory is controlled based on whether the pass/fail judgment is to be made for all areas of the memory cells arranged in a matrix, or for which area of the memory cells. It outputs an access position control signal to the memory control section 7 for performing the following.

The memory control unit 7 supplies various clock signals and read signals to the X address counter 4, the Y address counter 5, and the XY projection circuit 11 based on the access position control signal from the area control unit 6 and the control signal from the processor device 12.
It outputs a command signal for commanding read (read) or write (write).

The binarization circuit 8 operates based on a preset level value A and value when the value of image data for each pixel displayed in 8 bits stored in each memory cell is in the range of A≦D≦B. is associated with 1, and when the value of the image data is other than that, 0 is associated, and the binarization process is performed and output.

The binarization level A setting means 9 and the binarization level B setting means 10 are for setting the level values A and B used in the binarization circuit 8, respectively.

The XY projection circuit 11 adds the binarized data for each row and column of the matrix image memory based on the binarized data and the (:ARRY signal) of the X address output from the memory control unit 7. It is used for searching image data stored in the image memory 3.

FIG. 3 shows a case where an object such as a camera projects image data for the number "1"' by the XY projection circuit 11. The processor device 12 controls the entire circuit described above.

Next, the operation of the image data quality determination method according to this embodiment will be explained based on FIGS. 2 to 5.

In order to perform a pass/fail judgment according to this embodiment, image data for pass/fail judgment is transferred from the processor device and temporarily written into the image memory 3 in step SJI. As shown in FIG. 5, the image data for quality determination is, for example, 8-bit image data in which each pixel has data of either 55 or AA in hexadecimal notation arranged in each pixel. In this embodiment, the quality determination is made for, for example, a 7×7 matrix area as shown in FIG. 5 among all areas of the image memory.

In step SJ2, the binarization level A setting means 9 sets the binarization levels A and B used in the binarization circuit 8.
and by the binarization level B setting means 1°. In this embodiment, for example, A is set as A=AA, and B is set as B≧
AA.

In step SJ3, reading of all rows and columns of the area where the X address and Y address are set is started from the image memory.

In step SJ4, the binarization circuit 3 converts the level A,
The image data read from the image memory based on B is binarized for each memory cell, that is, for each pixel. In this example, 0 is associated with the image data 55 shown in FIG. 5, and 1 is associated with the image data AA. The lower part of FIG. 5 shows a matrix whose matrix elements are binary data.

In step SJ5, the XY projection circuit 11 adds up the binarized data for rows and columns using an X counter and a Y counter.

Below and on the right side of the matrix in the lower row of Figure 5 are the binarized data X for each row and column of the matrix.
The addition results added by the Y projection circuit 11 are shown.

In step SJ6, the image data for quality determination read out from the image memory 3 is subjected to binarization processing and the added data is binarized, and the image data for quality determination before being stored in the image memory 3 is subjected to the same binarization process. A comparison is made with the processed and added data for the area targeted for quality determination. In this example, the pass/fail judgment is made on a 7x7 matrix, so the total result of addition of rows and columns is only 7+7=14, and unlike the conventional method, 7x7.49 pieces of image data are added for the entire area. There is no need to perform verification.

That is, since the amount of data to be compared is small, the burden required for comparison processing on the processor device is significantly reduced, and the processing time is also significantly shortened.

If it is detected in step SJ7 that a point of disagreement has occurred, the point of disagreement is extracted and displayed as an error point in step SJ9, and the processing of the method is ended.

For example, if it is detected that the value of the addition result does not match the value obtained in advance in the third row and fourth column shown in FIG. There is an abnormality in the cell, and the location of the memory cell (3, 4)
) is displayed.

On the other hand, if there is no mismatch, the process may proceed to step SJ2 via step SJ8, and the same procedure may be repeated by changing the values of A and B of the binarization level, or the same procedure may be repeated by changing the judgment area. It is also possible to repeat the procedure to determine pass/fail for all areas.

In this manner, in this example, it is possible to determine the quality of the required area for the entire area or a specific area of the image memory without using the bus of the processor device, thereby reducing the processing time.

(Effects of the Invention) The present invention uses a new dedicated hard logic in terms of circuitry by using an It is possible to easily judge whether the product is good or bad without having to do so. Furthermore, in the method according to the present invention, there is no need to use a processor to check all cells of the memory one by one, and since an XY projection circuit is used, the response time of the processor is not required, and only the memory access time is required. This will shorten the quality determination processing time.

Furthermore, since the quality of the image memory can be determined with almost no dependence on the processing of the processor, the load on the bus of the processor can be reduced.

[Brief explanation of the drawing]

Fig. 1 is a flowchart of the principle of the present invention, Fig. 2 is a circuit diagram of an embodiment of the invention, Fig. 3 is an explanatory diagram of XY projection, Fig. 4 is a flowchart of the embodiment, and Fig. 5 is a pass/fail judgment. FIG. 3 is a diagram showing an example of image data and binarized data. Sl...Writing image data for pass/fail judgment to image memory S2...Binarization processing

Claims (1)

[Claims] Predetermined image data for quality determination is stored in an image memory in which memory cells for storing image data in units of a predetermined word length are arranged in a matrix (S1) to determine the quality of the image memory. In this method, the once stored image data for determination is read out, and the data is subjected to certain arithmetic processing for each memory cell and binarized (S2). Addition processing is performed on the data for each row and column of the matrix using an XY projection circuit (S3
), the quality of the image memory is determined by comparing the addition result with the result obtained by the same process as the process for the determination image data before being stored in the image memory (S4). Characteristic image memory quality determination method.