JPH0962763A - Bar code detection method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To specify a bar code and to detect the position and the tilt of the bar code even when the position, size and tilt of the bar code are unspecified by detecting the connection area where the feature value of an acquired shape shows a rectangle and also many graphics are included in the same tilt direction as a bar code area. SOLUTION: The images inputted from an image input device 8 are stored in an image memory 5 and converted into the binary data. A connection area consisting of black pixels more than a prescribed number is extracted. Then an AND is calculated between the connection area and the binary data, and the graphics including black pixels more than a prescribed number are extracted. The tilt of a main axis is calculated against every graphic, and the number of graphics of the same direction are totalized in every connection area based on the axis tilt calculated against every graphic. As the bars of a bar code have the same axis tilt, the connection area including many graphics of the same tilt is defined as a bar code area. As a result, the bar code area is detected as a connection area that includes many graphics of the same shape and same tilt.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a barcode detecting method for detecting the position and inclination of a barcode from an image of an article input from an image input device, and more particularly to an article automatically conveyed by a belt conveyor or the like. The present invention relates to a barcode detection method for detecting the position and inclination of a barcode in order to determine the optimum scanning pattern of the barcode scanner for each article when the attached barcode is automatically read by the barcode scanner.

[0002]

2. Description of the Related Art As a technique for extracting a barcode area from an image, there are disclosed in Japanese Patent Laid-Open Nos.
No. 324896 is known. JP-A-2-125
In Japanese Patent No. 381, the input image signal is binarized and stored in a memory, and the binary data is converted into contour data.
The outline data is expanded (enlarged) to fill the barcode area, and the output data after expansion processing is reduced or shrunk to a size that does not erase the filled barcode area to erase the background image. Then, after calculating the slope and center point of the remaining barcode area data,
Data is read from the memory in the tilt angle direction to decode the bar code.

Japanese Patent Laid-Open No. 5-324896 discloses
Performs thickening processing to convert adjacent white pixels into black pixels for consecutive black pixels in the image, obtains the area of each black pixel region connected in the thickened processed image, and calculates the area of these black pixel regions. The maximum area is recognized as a barcode symbol area, the logical product of each pixel is calculated for the barcode symbol area and the original image in the processed image, and the barcode is identified from the obtained black pixels.

[0004]

DISCLOSURE OF THE INVENTION Problems to be Solved by the Invention
According to the publication No. 1, in order to set the enlargement process for filling the barcode region and the reduction / contraction process for removing the background pixel, the bar width of the barcode and the size of the region are known in advance. Must be Therefore, there is a problem that it cannot be applied when the size of the barcode is unspecified.

Further, in Japanese Patent Laid-Open No. 5-324896, the approximate position of the bar code is known in advance, and this is a method of identification when a narrow area including the bar code is the target image, and the image is picked up. When the target area is expanded, the figure area other than the barcode is also included, and the probability that the thickened black pixel area having the largest area becomes the barcode area is low. Therefore, there is a problem that it cannot be applied when the position of the bar code is unspecified and the imaging target area including the bar code is wide.

Therefore, according to the present invention, in the image of the entire article including the bar code area, even if the bar code position, size and inclination are unspecified, the bar code can be specified to detect the bar code position and inclination. A barcode detection method is provided.

[0007]

According to the first aspect of the present invention,
An article area including a barcode is extracted from an image input from an image input device, the image of the article area is converted into binary data and stored, and the binary data is expanded and contracted a predetermined number of times to connect black pixels. A region is created, a plurality of connected regions composed of a predetermined number or more of black pixels are extracted from the created connected region, a feature value of the shape is obtained for each extracted connected region, and each extracted connected region is extracted. A logical product of the area and the binary data is obtained to extract a figure in which a predetermined number or more of black pixels are continuous in each connected area, the inclination direction is obtained for the extracted figure, and the figure is classified according to the inclination for each connected area. The number of shapes is calculated, and the calculated feature quantity and the number of graphics by inclination are compared for each connected area.The connected area where the shape feature shows a rectangle and the number of figures in the same tilt direction is large is the barcode area. To detect A.

[0008]

Embodiments of the present invention will be described below with reference to the drawings. In FIG. 1, reference numeral 1 denotes a CPU (central processing unit) that constitutes the main body of the control unit, and to this CPU 1 a ROM (Read Only Memory) 3, a RAM (Random Access Memory) 4, an image via a bus line 2. A memory 5, a video I / F (interface) 6 and a peripheral device I / F (interface) 7 are connected to each other. An image input device 8 such as a camera is connected to the video I / F 6. The CPU 1 controls each unit based on the program data stored in the ROM 3. The peripheral device I / F 7 is connected to a bar code scanner and various devices that drive the scanner.

The image input device 8 captures an image of a bar code-added article and outputs an image signal. The image signal from the image input device 8 is stored in the image memory 5 as an input image via the video I / F 6. The image memory 5 stores in advance a background image in a state in which no article is present.
As shown in FIG. 2, the image information stored in is the input image IM1 and the background image IM2.

When the CPU 1 stores the input image in the image memory 5, the CPU 1 performs a process based on the flowchart of FIG.
First, in S1, the article area including the barcode is extracted from the input image of the image memory 5. That is, the input image IM
The difference value between 1 and the background image IM2 is calculated, and the area surrounded by the pixels having the difference value equal to or larger than the threshold value is extracted as the article area.

Subsequently, in S2, the extracted article image of the article area is converted into binary data by a predetermined threshold value and stored in the image memory 5. Then, in S3, the stored binary data is enlarged and shrunk a predetermined number of times to create a black pixel connected region, and a connected region composed of a predetermined number or more of black pixels is extracted. FIG. 4 shows a result of performing the enlarging and degenerating operations a predetermined number of times on the binary data of the input image IM1 of FIG. 2, and nine connected regions a to i are created as a whole, and the bars The code region is included as the connecting region i.

Subsequently, in S4, the feature quantity of the shape for each of the connected regions a to i is obtained. The feature amount of the shape includes circularity (= 4 × π × area / (perimeter) ² ) or irregularity (1 / circularity), moment coefficient, and the like, which are obtained by the area and the perimeter. Further, in S5, a logical product of the connected area (connection coefficient) and the binary data stored in the image memory 5 is obtained, and a figure in which a predetermined number or more of black pixels are continuous is extracted. FIG. 5 shows an image of the extracted figure, which is shown in FIG.
Although the entire value data (characters and bar codes) is extracted as it is, actually, by limiting to a figure in which a predetermined number or more of black pixels are continuous, there is an effect of removing an isolated point such as noise.

Next, in S6, the direction of inclination (inclination of the main axis) with respect to the figure is obtained, and the number of figures is summed up for each connected region in each of the predetermined directions. That is, the number of figures in the same direction is totaled from the inclination of the main axis obtained for each figure for each connected area. Reference numeral 11 in FIG. 5 shows an example of the inclination of the main axis with respect to the numeral. The inclination of the main axis is not constant but differs for each number. On the other hand, in the case of a barcode, the inclination of the main axis of each bar is the same.

Subsequently, in S7, the feature amount of the shape of the connected area and the number of distinct tilts of the graphic are compared for each connected area. Then, in S8, a connected area having a large number of figures having a rectangular shape and the same inclination is detected as a barcode area. In addition,
In detecting the barcode area, it is possible to receive the result of the comparison for each connected area and rank the areas having the highest probability as candidates for the barcode area.

Information on the position and inclination of the thus detected bar code area is stored in the RAM 4, and the peripheral area I / F 7 is read out as needed by controlling the focal length of the bar code scanner and the scanning pattern. It is used to control various devices through.

In such a configuration, the input image from the image input device 8 is stored in the image memory 5, the difference value between the input image and the background image is obtained from the stored information in the image memory 5, and the difference value equal to or greater than the threshold value is obtained. An area surrounded by pixels having is extracted as an article area. That is, the input image IM of FIG.
1 is extracted. Then, the input image IM1 is converted into binary data, and the binary data is enlarged and shrunk a predetermined number of times to create a black pixel connected area, and a connected area composed of a predetermined number or more of black pixels. Is extracted.

Further, a logical product of the connected area and the binary data is obtained, and a figure in which a predetermined number or more of black pixels are continuous is extracted. Then, the inclination of the main axis is obtained for each figure, and the number of figures in the same direction is totalized from the inclination of the main axis obtained for each figure for each connected region. Since the bars of the barcode all have the same inclination of the main axis, a connected area where the number of figures having the same inclination is large is a barcode area. Therefore, the barcode area is detected as a connected area having a large number of figures having a rectangular shape and the same inclination.

Since the information indicating the position and inclination of the bar code area is obtained in this manner, this information is used as the peripheral device I / I.
By being supplied to the drive unit of the barcode scanner via F7, the barcode scanner can perform the optimum scanning for the barcode added to the article and automatically read the barcode.

Further, by performing such processing, even if the position, size and inclination of the barcode in the image are unspecified, it is possible to reliably identify the barcode and obtain information indicating the position and inclination. .

In the above-mentioned example, the case where the bar code is created as one connected area has been described. However, the bar code may be large or the bar code may be enlarged or shrunk depending on the number of times of the binary data. May be divided into a plurality of connected areas. For example, if the number of times of enlargement / reduction is reduced, the connected area i of the barcode area is divided into connected areas i1 and i2 as shown in FIG. The connection area e is also divided into e1 to e4, and the connection area f is also divided into f1 to f3.

When divided in this way, the connected region i1
Since there are a large number of figures having the same inclination in the same direction, the number is detected as a barcode by the processing of FIG. In this case, when the regions having the highest probability as the barcode region candidates are ranked according to the result of the comparison for each connected region, the connected region i1 is detected as the first candidate and the connected region i2 is detected as the second candidate. To be done. The detected tilt information of the bar code is not different between the connected area i1 and the connected area i2, and the position information is only slightly shifted. Moreover, the connection area i1 and the connection area i2 are adjacent to each other. Therefore, the bar code scanner can scan the connection area i1 and the connection area i2, that is, the bar code without any problem.

[0022]

As described above, according to the first aspect of the invention, in the image of the entire article including the barcode area, the barcode is specified even if the position, size and inclination of the barcode are unspecified. The position and inclination of the barcode can be detected.
Therefore, it is possible to automatically read the barcode from the article using the barcode scanner.

[Brief description of drawings]

FIG. 1 is a block diagram showing a configuration of an embodiment of the present invention.

FIG. 2 is a diagram showing an example of image information stored in an image memory according to the same embodiment.

FIG. 3 is a flowchart showing a barcode detection process by the CPU in the same embodiment.

FIG. 4 is a diagram showing a result of enlarging and reducing an input image in the same embodiment.

FIG. 5 is a view showing a figure in which black pixels extracted in the embodiment are consecutive.

FIG. 6 is a diagram showing another result of enlarging and reducing an input image in the same embodiment.

[Explanation of symbols]

 1 ... CPU (Central Processing Unit) 5 ... Image Memory 8 ... Image Input Device

Claims (1)

[Claims] 1. An article area including a bar code is extracted from an image input from an image input device, the image of the article area is converted into binary data and stored, and the binary data is expanded and contracted a predetermined number of times. Create a connected area of black pixels, extract a plurality of connected areas composed of a predetermined number or more of black pixels from the created connected area, and obtain a shape feature amount for each of the extracted connected areas. , A logical product of each extracted connected area and binary data is obtained to extract a figure in which a predetermined number or more of black pixels are continuous in each connected area, and an inclination direction is obtained for the extracted figure to obtain each connected area. The number of figures for each slope is totaled, and the calculated feature quantity and the number of figures for each tilt are compared for each connected area, and the shape feature indicates a rectangle and there are many connected areas in the same tilt direction. As the barcode area Barcode detection method and detecting.