JPS6041174A - Processing method of picture data - Google Patents

Abstract

PURPOSE:To obtain a picture data processing method evaluating the relation between a picture and a certain position by selecting a weighted value matrix obtained by weighting a certain position with a picture data pattern having a same matrix configuration. CONSTITUTION:The data of the weighted value matrix B and picture data A are stored in a memory M1 and memory M2 respectively by using the matric memories M1-M3 having a mXn cell configuration. A gate circuit G is opened/closed by a true or false value of ''1'' or ''0'' every cell in accordance with the contents of the memory M2 to control whether the data of the memory M1 is to be transferred to a memory M3 or not. In this case, the data are read out in parallel and parallel processing is executed instead of the serial and sequential processing of picture data to obtain a high speed processing.

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to an image data processing method for evaluating the relative relationship between an image and a certain position of interest.

Prior Art Conventionally, for example, as preprocessing for pattern recognition, image data of a pattern read dot by dot by a scanner is processed by a computer to evaluate the relative relationship between the image and a certain position of interest. One screen worth of image data is hundreds of thousands of dots from the enemy, and a general computer requires one instruction execution time to process each dot, resulting in a very large amount of execution time overall. It's gone.

Purpose The present invention has been made in consideration of the above points, and provides an image data processing method that makes it possible to determine the relative relationship between an image and a certain position of interest.

An embodiment of the present invention will be described below in detail. First, image data processing will be described below in order to evaluate the relative relationship between one image and a certain position of interest.

For example, as shown in Figure 1, the 9x9 to be processed
When processing image data A of a pattern with a dot configuration, a calculation using a weight value matrix B weighted in inverse proportion to the distance is performed using a 5×5 area at the center of the matrix with the same dot configuration as the point of interest. The processing is performed and the processing result C is obtained. The content of the calculation process is to use the matrix of image data A as a mask and set the values of the weight value matrix B corresponding to the "b" dots in the pattern. As a result of the processing result C, a value representing the relative relationship between the pattern of the image data A and the position of interest is obtained.

FIG. 2 shows an expanded version of the process shown in FIG. 1, in which the result A obtained by performing a logical operation on the basic pattern matrix D and the image data A is used as a mask. The logical operations include AND, OR,
Various logical operations such as EX-OR are applicable.

This means of processing image data A through unweighted weight value matrix B to obtain result C forms the basis of image processing, and the present invention particularly aims to perform such image data processing at high speed. It was designed to allow

Therefore, in the present invention, as shown in FIG. 3, matrix memories Ml, M2 . M3 is used to store the data of the weight value matrix B in the memory M1 and the image data A in the memory M2, and as shown in FIG. Its ll
The gate circuit G is opened and closed depending on the truth value of l and 10'', and control is performed as to whether or not to transfer the data in the memory M1 to the memory M3. Instead of serially performing data processing, high-speed processing is possible by reading the data in parallel and executing parallel processing. Figure 5 shows the processing flow for sequential processing. 6
The figure shows the processing flow for parallel processing. In the figure, MASK (i, j) indicates image data A, VAL (i, j) indicates processing result C data,
PATH(i, j) indicates data of weight value matrix B, respectively. In addition, when executing the process shown in FIG. 2, the process shown in FIG. 7 is added to each X portion in FIGS. 5 and 6. In the figure, MASKI (i.

j) means the basic pattern matrix D, MAS
K2 (i, j) is image data A.

MASK3 (i, j) respectively indicate image data A' that has been subjected to logical processing.

As described above, in the image data processing method according to the present invention, the relative relationship between an image and a certain position of interest is determined by selecting a matrix value weighted with respect to a certain position using a pattern of image data as a mask. It has the excellent advantage that processing results for evaluating relationships can be immediately obtained, and basic operations in bitmap-structured digital image processing can be performed in a general-purpose manner.

[Brief explanation of the drawing]

1 and 2 are diagrams each showing an example of image data processing according to the present invention, FIG. 3 is a diagram showing a matrix structure used when concretely implementing the present invention, and FIG. A block diagram showing an example of the control means, FIGS. 5 and 6 are diagrams each showing an example of the flow of image processing according to the present invention, and FIG. 7 is a flow diagram added when executing the process in FIG. 2. FIG. A: Image data B: Weight value matrix C: Processing result D: Basic pattern G: Gate circuit Figure 3, Figure 4 Figure 5

Claims (1)

[Claims] Image data in which a result of evaluating the relationship between an image and a certain position of interest is obtained by selecting a weight value matrix weighted with respect to a certain position according to a pattern of image data having the same matrix configuration. Processing method.