JPH0448277B2 - - Google Patents

Description

[Detailed Description of the Invention] [Summary] Processing the density levels of subdivided pixels as individual data requires a huge storage capacity and long processing time to store dictionary data and data to be processed. do. Therefore, data is compressed to reduce storage capacity and processing time.

[Industrial application field]

The present invention relates to an image processing apparatus using an electronic computer, and more particularly to an image data processing method that enables reduction of storage capacity and processing time.

2. Description of the Related Art Image processing devices that use electronic computers to compare arbitrary images with reference images stored in the device are being developed. As a means for representing a binarized image in such an image processing apparatus, a method is generally adopted in which the density levels of pixels obtained by subdividing an image are listed in accordance with the scanning order. However, as the precision of image data increases, the number of pixels increases, requiring a huge capacity storage device to store dictionary data and data to be processed, and a long processing time to process such data. I need.

Therefore, in order to improve the accuracy of image data in an image processing device, there is a desire to develop an image data processing method that can reduce storage capacity and processing time.

[Conventional technology]

FIG. 3 is a diagram showing a normal image data processing method, FIG. 3a is a diagram showing an example of pixels, and FIG. 3b is a diagram showing an example of image data.

In a normal image data processing method, a reference image and an image to be compared with it are subdivided according to the same standard, and the density level of each pixel is binarized and listed according to the scanning order. For example, in FIG. 3a, the image is divided into 10 parts in both the X and Y directions, forming 100 pixels. In Figure 3b, the density level of each pixel is binarized, and the white part is represented by "0" and the asterisk part is represented by "1".The density level of the pixel in Figure 3b is shown in the scanning order. Therefore, when listed, it becomes image data.

When processing an image, the density levels of the corresponding pixels of the reference image and the processed image to be compared with it are sequentially compared, and the number of pixels with different density levels in an arbitrary interval is counted, and the reference image and the processed image are compared. The amount of non-matching is calculated.

[Problem that the invention seeks to solve]

In the above example, the number of pixels is 100, and if there is a storage capacity of at least 200 bits, it is possible to store the dictionary data representing the reference image and the processed data representing the processed image. However, in the normal image data processing method, when the number of divisions of the interval is increased and the precision of the image data is increased, the number of pixels increases exponentially, and a huge amount of memory is required to store the dictionary data and the data to be processed. This method requires a large-capacity storage device and requires a long processing time to compare such huge amounts of data one by one.

[Means for solving problems]

FIG. 1 is a diagram showing the image data processing method according to the present invention, FIG. 1a is a diagram showing the structure of unit image data, and FIG. 1b is a diagram showing an example of the image data.

The above problem consists of a 1-bit density indicator bit 1 that indicates the density level, and a 7-bit length indicator bit 2 that indicates the continuous length of the density level.
Byte unit image data is listed according to the scanning order, and when processing the image, the dictionary data listed according to the above method and the data to be processed are compared, and the parts where the density indicator bit 1 differs in any interval are determined. This problem is solved by the image data processing method of the present invention, which calculates the amount of unmatching between the dictionary data and the data to be processed by sequentially adding the lengths.

[Effect]

In FIG. 1, by arranging 1-byte unit image data according to the running order, the storage capacity for storing dictionary data and processed data is reduced, and the dictionary data and processed data are not matched. The processing time for determining the amount can be shortened.

〔Example〕

Embodiments of the present invention will be described below with reference to the accompanying drawings. Note that FIG. 2 is a block diagram showing one embodiment of the image processing apparatus.

In FIG. 1a, 1-bit concentration indicator bit 1
By configuring 1 byte of unit image data with , and 7-bit length indication bit 2, it is possible to represent a section where the same density continues regardless of the number of divisions of the section with 1 byte of unit image data. . For example, if the content of FIG. 3b is expressed by unit image data according to the present invention, it will be replaced with 22 unit image data, that is, 172-bit image data, as shown in FIG. 1b, and the dictionary data and the data to be processed will be Requires at least 360 bits of storage capacity to store.

It goes without saying that the normal method is advantageous when the number of divided sections is small as shown in the figure. However, if you increase the number of divisions of the section by 20 times, the number of pixels will increase by 400 times,
While conventional methods require a storage capacity of 80,000 bits to store dictionary data and data to be processed, the method of the present invention requires 8,000 bits, which is about 20 times as large.
A storage capacity of bits is sufficient, and the storage capacity for storing dictionary data and data to be processed can be significantly reduced.

However, the method according to the present invention in image processing differs from normal methods in that it sequentially compares the density levels of corresponding pixels and counts the number of pixels with different density levels, thereby preventing non-matching between the reference image and the processed image. It is not possible to calculate the amount. Therefore, in the method of the present invention, the amount of unmatching between the reference image and the processed image is calculated by the following method.

In FIG. 2, the image processing device includes a register 31
and a register 32, and a non-matching amount addition circuit 4, unit image data A ( (referred to as dictionary data A) is stored in the register 31.
Also, unit image data B (referred to as processing data B) sequentially read out from the processed data is stored in the register 32.

() Calculation of non-matching amount: In the arithmetic circuit 3, the non-matching amount C is calculated from the dictionary data A and the data to be processed B using the length indication bit 2 according to equation (1).

C=A+B-|A-B|/2...(1) This non-matching amount C is input to the non-matching amount adding circuit 4, but the non-matching amount adding circuit 4
is configured to operate and add the non-matching amount only when the density indicator bit 1 of the dictionary data A and the processed data B is different, and it is possible to add the non-matching amount in any section where the density indicator bits are different. By sequentially adding the lengths, the amount of unmatching between the dictionary data and the data to be processed can be determined.

() Length verification: Also, the arithmetic circuit 3 calculates the difference in length between the dictionary data A and the data to be processed B from the length indication bit 2 using equation (2), and stores the result in the register 31. , is used as the next dictionary data A to perform the above-mentioned unmatching amount calculation.

A'=|A-B|+A-B/2...(2) Also, calculate the difference in length between the dictionary data A and the data to be processed B using equation (3), and send the result to the register 32.
, and use it as the next dictionary data B to perform the above-mentioned unmatching amount calculation.

B'=|A-B|-A+B/2...(3) If A' stored in register 31 or B' stored in register 32 is 0, the next dictionary data A is stored in register 31 or Processing data B is stored in the register 32.

() Example of calculating unmatched amount: Dictionary data A and processed data B shown in the example in the table below.
Calculate the amount of unmatching in .

Example of dictionary data A and processed data B Concentration level: 0 1 0 Length of dictionary data A: 60, 80, 60, Length of processed data B: 50, 100, 50, 1st operation: Calculation of non-matching amount is omitted as it is at the same level.

A′=10, B′=0, 2nd operation: C=10, A'=10, B'=90 3rd operation: Calculation of non-matching amount is omitted as it is at the same level.

A′=0, B′=10, 4th operation: C=10, A'=50, B'=0 5th operation: Calculation of non-matching amount is omitted as it is at the same level.

A'=0, B'=0 As described above, the dictionary data A and the processed data B are compared through the five calculations, and it is calculated that the amount of unmatching within that section is 20.

Comparing the above examples of dictionary data A and processed data B using the usual method and calculating the amount of unmatching results in
Dictionary data A for each of the 200 pixels
and the data to be processed B, and if the density level differs, it is added each time, which requires a long processing time. That is, the method according to the present invention can significantly reduce the time required for image processing.

〔Effect of the invention〕

As described above, according to the present invention, it is possible to provide an image data processing method that can reduce storage capacity and processing time.

[Brief explanation of drawings]

FIG. 1 is a diagram showing an image data processing method according to the present invention, FIG. 2 is a block diagram showing an embodiment of an image processing device, and FIG. 3 is a diagram showing a normal image data processing method. . In the figure, 1 is a density indicating bit, 2 is a length indicating bit, 3 is an arithmetic circuit, and 4 is a length indicating bit.
3 represents a non-matching amount addition circuit, and 31 and 32 represent registers, respectively.

Claims (1)

[Claims] 1. As a means for representing a binarized image in an image processing device, a 1-bit density indicator bit 1 indicating a density level and a 7-bit length indicator bit indicating a continuous length of the density level are used. 2, 1-byte unit image data is listed according to the scanning order, and when processing the image, the dictionary data listed according to the above means is compared with the data to be processed, and the density indication bit 1 is set in an arbitrary interval. 1. A method for processing image data, characterized in that the amount of unmatching between dictionary data and data to be processed is determined by sequentially adding the lengths of portions where the values are different.