CN101320471A - image analysis method - Google Patents

Abstract

The present invention relates to an image analysis method. An image block arrangement table is constructed by inputting coordinates of a plurality of image blocks included in an image and analyzing the image according to the image block arrangement table. The image analysis method provided by the invention can expand the image block arrangement list at any time, and the types of images which can be analyzed by an image analysis program can be greatly increased on the premise of sufficient memory, thereby being quite convenient.

Description

image analysis method

technical field

The invention relates to an image analysis method, in particular to an analysis method capable of analyzing images with different arrangements of image blocks.

Background technique

Devices for digital image processing, such as digital cameras, document scanners, printers, etc., have been widely used by people today. In order to ensure that the output image produced by the digital imaging device is as consistent as possible with the real image, for example, whether the image printed by the printer is very close to the electronic image, or whether the electronic image produced by the scanner is close to the paper image, so it is necessary Some image analysis steps to distinguish the difference between the output image and the real image. This kind of image analysis process is generally the easiest to identify by human eyes, that is, the difference between the output image and the original image is directly compared by human eyes. However, such a manual identification method requires a considerable amount of manpower and time, and the identification by human eyes is biased towards personal subjective judgment, which is not an objective method. Therefore, it is known that a process of performing image analysis using an image analysis program has been developed.

Generally, the image to be analyzed is composed of multiple blocks, and each block is a picture image or a text image. Generally speaking, a single known image analysis program can only analyze a single image with a specific arrangement of image blocks, but cannot be applied to another image with a different arrangement of image blocks. Please refer to FIG. 1(a) to FIG. 1(c), which are schematic diagrams of images with different block arrangements. Fig. 1 (a) has a first image IA, Fig. 1 (b) has a second image IB, and Fig. 1 (c) has a third image IC, the first image IA includes five blocks 1-5, the second The image IB is the rotated image of the first image IA, and the third image IC is obtained by rearranging the positions of the five blocks in the first image IA.

Please refer to FIG. 2( a ), which is a schematic diagram of a preferred embodiment of a known image analysis method. In Fig. 2 (a), the first image analysis program PA is specially designed for the first image IA, the second image analysis program PB is specially designed for the second image IB, and the third image analysis program PC is specially designed for the first image IA. designed for the third image IC. Please refer to FIG. 1(a) to FIG. 1(c) and FIG. 2 at the same time. The program writer writes the position information of the five blocks of the first image IA in the first image analysis program PA, so the first image analysis program The PA can analyze the first image IA according to the location information of the blocks. Similarly, the second image analysis program PB can analyze the second image IB, and the third image analysis program PC is used to analyze the third image IC.

Although the second image IB is only obtained by rotating the first image IA, as shown in Figure 1(a) to Figure 1(c), if the program writer of the first image analysis program PA did not consider In the case of image rotation, the first image analysis program PA can only analyze the first image IA but cannot analyze the second image IB. In this case, if you want to analyze the second image IB, you must write a second image analysis program PB for the second image IB, or modify the first image analysis program PA. However, the process of modifying the first image analysis program PA is not as simple and fast as writing the second image analysis program PB.

If the program writer of the first image analysis program PA considers the image rotation and adds the image rotation command RC into the first image analysis program PA when writing the program, then the first image analysis program PA can analyze the first The image IA and the rotated first image PA (ie, the second image IB) are shown in FIG. 2( b ). However, the third image IC that has been rearranged by blocks still cannot be analyzed by the first image analysis program PA. At this time, a third image analysis program for analyzing the third image IC must be designed for the block position of the third image IC. PC.

In other words, for image analysis programs, images with different arrangements of image blocks will be regarded as different images and thus cannot share the same image analysis program. Even the image analysis program PA added with the image rotation command RC as shown in Figure 2(b) can only analyze the second image PB obtained after the rotation of the first image PA, but still cannot analyze the third image PC or other more images with different arrangements of blocks.

It is cumbersome and inefficient to have to rewrite the program to analyze the image every time an image with a block arrangement appears, and therefore, there is a need for an image that can analyze images with different image block arrangements using a single analysis program Analytical method.

Contents of the invention

The purpose of the present invention is to provide an image analysis method, especially an image analysis method that can analyze images with different image block arrangements according to the image block arrangement table.

In a preferred embodiment, the present invention proposes an image analysis method for analyzing multiple images with different block arrangements, wherein each of the multiple images includes multiple image blocks and each block has Starting point coordinates and end point coordinates, this method includes:

input the start coordinates and end coordinates of multiple blocks contained in each image;

recording the start coordinates and end coordinates of multiple blocks included in each image in the image block arrangement table; and

Multiple images are analyzed according to the image block permutation table.

In a preferred embodiment, the plurality of blocks are picture images.

In a preferred embodiment, the plurality of blocks are text images.

In a preferred embodiment, the multiple blocks include pictures and text images.

The invention not only improves the image analysis method, but also expands the image block arrangement table at any time. Under the premise of sufficient memory, the types of images that can be analyzed by the image analysis program can be greatly increased, which is quite convenient.

Description of drawings

FIG. 1( a ) to FIG. 1( c ) are schematic diagrams of images with different block arrangements.

2(a) and 2(b) are schematic diagrams of a preferred embodiment of a known image analysis method.

Fig. 3 is a schematic diagram of an image block arrangement table of a preferred embodiment of the image analysis method of the present invention.

Fig. 4 is a flowchart of a preferred embodiment of the image analysis method of the present invention.

Fig. 5 is a schematic diagram of an image block arrangement table of a preferred embodiment of the image analysis method of the present invention.

Wherein, the reference signs are explained as follows:

1, 2, 3, 4, 5 blocks

401, 402, 403 steps

IA, IB, IC image

PA, PB, PC, PD image analysis program

T image block arrangement table

R5 block radius

X ₀₀ , Y ₀₀ , X _nn , Y _nn image coordinates

X ₁₁ , X ₂₁ , X ₃₁ , X ₄₁ , X ₅₁ , X _1n , X _2n , X _3n , X _4n , X' ₁₁ , X' ₂₁ , X' ₃₁ , X' ₄₁ , X' ₅₁ , X' _1n , X' _2n , X' _3n , X' _{4n ,} Y ₁₁ , Y ₂₁ , Y ₃₁ , Y ₄₁ , Y ₅₁ , Y _{1n , Y 2n ,} Y 3n , Y _4n , Y' ₁₁ , Y' ₂₁ , Y' ₃₁ , Y' ₄₁ , Y' ₅₁ , Y' _1n , Y' _2n , Y' _3n , Y' _4n block coordinates

Detailed ways

In order to prevent the image analysis program from being limited to analyzing images with a specific image block arrangement, the present invention proposes an image block arrangement table. The image block arrangement table records position information of multiple image blocks of the image to be analyzed, so that a single analysis program can analyze images with different block arrangements according to the information in the image block arrangement table. As shown in FIG. 3 , the fourth image analysis program PD can analyze the first image IA, the second image IB and the third image IC by reading the image block arrangement table T of the present invention.

Please refer to FIG. 4 , which is a flowchart of a preferred embodiment of the image analysis method of the present invention. In FIG. 4 , the image analysis starts, and in step 401 , the starting point coordinates and the ending point coordinates of each image block of the image are input. In step 402, record the starting point coordinates and the ending point coordinates in the image block arrangement table T. Then in step 403 the image is analyzed according to the coordinate data of the image blocks recorded in the image block arrangement table T.

In the image analysis method of the present invention, the position information of each image block included in the image is recorded in the image block arrangement table T, so that the image analysis program can obtain the arrangement information of the image blocks from the image block arrangement table and then proceed analyze.

Please refer to FIG. 5 , which is a schematic diagram of an embodiment of the image block arrangement table T of the present invention. The data stored in the image block arrangement table T of the present invention includes starting point coordinates and end point coordinates. If the shape of the block is a circle, store the center and radius of the circle. If the same image block is divided into multiple small blocks, the number of divided lengths and widths of the image block is stored. Of course, taking common rectangular blocks and circular blocks as examples does not limit the present invention to only analyze image blocks of these two block shapes.

Please refer to FIG. 1( a ) to FIG. 1( c ) and FIG. 5 at the same time to illustrate the flow of the image analysis method of the present invention again. In Fig. 1 (a) to Fig. 1 (c), the first image IA includes a total of five blocks 1, 2, 3, 4 and 5 Blocks, wherein the first block 1 and the fifth block 5 are picture images, the second block 2 and the fourth block 4 are text images, and the third block 3 is a block with small block division. Define the point in the upper left corner in the first image IA as the origin, and its coordinates are (X ₀₀ , Y ₀₀ ), and define the point in the lower right corner in the first image IA as the end point, and its coordinates are (X _nn , Y _nn ) . The user can use the origin coordinates or the end coordinates as a reference point to know the start coordinates and the end coordinates of each image block. At this time, the user inputs the location information of the blocks into the image block arrangement table T respectively. The block information in this example includes: the first starting point coordinates (X ₁₁ , Y ₁₁ ) and the first end point coordinates (X _1n , Y _1n ) of the first block 1, the second starting point coordinates of the first block 2 (X ₂₁ , Y ₂₁ ) and the second end point coordinates (X _2n , Y _2n ), the third starting point coordinates (X ₃₁ , Y ₃₁ ) and the third end point coordinates (X _3n , Y _3n ) of the third block 3, The fourth starting point coordinates (X ₄₁ , Y ₄₁ ) and the fourth end point coordinates (X _4n , Y _4n ) of the fourth block 4, the fifth circle center coordinates (X ₅₁ , Y ₅₁ ) and the fifth Radius R5. When inputting the coordinates of the third block 3, the long side of the input block is divided into 5 pieces, and the wide side of the block is divided into 3 pieces, which means that the third block 3 is divided into 15 pieces in total. information. In addition, when the input end point coordinate is a single value, it is determined that the value is a radius, which means that the block is a circle. The above is the first way to manually input coordinate information. Of course, other ways of inputting coordinate information are possible. For example, please refer to the first image IA in Fig. 1(a), similarly define a point in the upper left corner of the first image IA as the origin of the image, its coordinates are (X ₀₀ , Y ₀₀ ), and define the first image The bottom right point in the IA is the end point of the image, and its coordinates are (X _nn , Y _nn ). When inputting coordinate information, the user uses the mouse to click on the block starting point of the first block 1, press and hold the left mouse button, drag the mouse cursor to the block end point of the first block 1, and then release it Click the left button of the mouse, and the program will know the first starting point coordinates (X ₁₁ , Y ₁₁ ) and first end point coordinates (X _1n , Y _1n ) of the first block 1 according to the image origin coordinates and image end point coordinates, and so on . When encountering the fifth circular block, use the mouse button to click on the center of the fifth block and drag it to its circumference, then release the button, and the program will provide a menu for the user to select the circled block Is rectangular or circular. After the block coordinates are input, the coordinate information is stored in the image block arrangement table T, as shown in FIG. 5 . When the fourth image analysis program PD wants to analyze the first image IA, it reads the coordinate data of the blocks of the first image IA from the image block arrangement table T to analyze the first image IA. When the fourth image analysis program PD wants to analyze the second image IB or the third image IC, the fourth image analysis program PD can analyze the second image IB without using the image rotation command RC, because FIG. 1( a) The image origin coordinates of the first image IA and the second image IB in Figure 1(b) are both (X ₀₀ , Y ₀₀ ), although the second image IB is a rotated image, as long as it is known as a reference The position of the image origin of the point, and the relative position between the origin and each block will not change due to rotation, so the coordinates of each block can be known, so in the image block arrangement table T, the coordinate information of the first image IA is exactly the same as the second image IB, as shown in FIG. 5 . The third image IC obtained by rearranging the block positions of the first image IA is regarded as a different image from the first image IA. Before analyzing the third image IC, the user also inputs data such as block coordinates of the third image IC, and the coordinate information of the third image IC is as follows: the first starting point coordinates of the first block 1 (X' ₁₁ , Y' ₁₁ ) and the first end point coordinates (X' _1n , Y' _1n ), the second starting point coordinates (X' ₂₁ , Y' ₂₁ ) and the second end point coordinates (X' _2n , Y' _2n ) of the first block 2 , the third starting point coordinates (X' ₃₁ , Y' ₃₁ ) and the third end point coordinates (X' _3n , Y' _3n ) of the third block 3, the fourth starting point coordinates (X' ₄₁ , Y' ₄₁ ) and the fourth end point coordinates (X' _4n , Y' _4n ), the fifth circle center coordinates (X' ₅₁ , Y' ₅₁ ) of the fifth block 5 and the fifth radius R5. These data are still stored in the image block arrangement table T. In the coordinate information of the third image IC, the value of the fifth radius is still R5 unchanged, but the coordinates of other points are changed due to the change of the block arrangement, as shown in FIG. 5 . When the fourth image analysis program PD can read the data of the image block arrangement table T to analyze the third image IC.

Similarly, for analyzing other images with different block arrangements, information such as block coordinates is firstly input, so that after the block position information is stored, the analysis program can directly analyze the image according to the block position information.

In summary, the present invention provides an image block arrangement table, so that images with different block arrangements can be analyzed by image analysis programs, without writing different analysis programs for images with different block arrangements. Of course, the image block arrangement table can add block position information of different images at any time, and is not limited to the three images in FIG. 5 . The invention not only improves the image analysis method, but also expands the image block arrangement table at any time. Under the premise of sufficient memory, the types of images that can be analyzed by the image analysis program can be greatly increased, which is quite convenient.

The above descriptions are only preferred embodiments of the present invention, and are not intended to limit the claims of the present invention. Therefore, all other equivalent changes or modifications that do not deviate from the spirit disclosed in the present invention should be included in the scope of this case. within the claims.

Claims (4)

1. An image analysis method for analyzing a plurality of images having different block arrangements, wherein each of the plurality of images includes a plurality of image blocks and each block has a start point coordinate and an end point coordinate, The methods include: Input the start coordinates and end coordinates of multiple blocks contained in each image; recording the start coordinates and end coordinates of multiple blocks included in each image in the image block arrangement table; and The plurality of images are analyzed according to the image block arrangement table. 2. The image analysis method according to claim 1, wherein the plurality of blocks are picture images. 3. The image analysis method according to claim 1, wherein the plurality of blocks are text images. 4. The image analysis method as claimed in claim 1, wherein the plurality of blocks include pictures and text images.

Images