JPH1091717A - Bar code recognizing device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a bar code recognizing device capable of quickly, stably and highly accurately recognizing a bar code by utilizing the differential picture of the bar code. SOLUTION: The picture information of a bar code written on an object 1 to be recognized is collected by a camera 2 and a photoelectric conversion part 3, a bar code area detection part 4 executes differential processing in the (x) direction or (y) direction of the collected picture information or in both the (x) and (y) directions, the threshold processing and edge detection processing of the differential processing result and feature extraction processing for the existing area of the bar code and a bar code recognizing part 5 executes the picture reconstitution of the bar code based on features in the existing area of the bar code.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a bar code recognition apparatus using a differential image.

[0002]

2. Description of the Related Art Conventionally, in a bar code recognizing apparatus for recognizing a bar code, a two-dimensional digital image of the bar code is collected, and when recognizing the bar code from the digital image, the bar code is recognized by some means. In general, a code candidate area is obtained, and recognition is performed using an image that has been binarized for the candidate area or an image that has been binarized from the image.

[0003]

However, in the case of the conventional apparatus described above, when performing recognition using a multi-valued image or a binary image, the capacity of an image memory for storing image data increases in recognition using a multi-valued image. However, there is a problem that the processing time increases.

[0004] In addition, high-speed processing is possible in recognition using a binary image, but when noise increases, the noise appears remarkably as quantization noise, and there is a problem in that stable and high-precision recognition is difficult.

SUMMARY OF THE INVENTION The present invention has been made in view of the above circumstances, and provides a barcode recognition apparatus that can realize stable, high-precision recognition in a short time by using a differential image of a barcode.

[0006]

A bar code recognition apparatus according to the present invention collects bar code image information described in an object to be recognized and a bar code information collection means. By performing a differentiation process on the image information in the x direction or the y direction or in both the x direction and the y direction, and performing a threshold process and an edge detection process on the result of the differentiation process, a feature extraction process on the barcode existence region is performed. A bar code area detecting means for performing the image reconstruction of the bar code based on the feature relating to the existence area of the bar code extracted by the bar code area detecting means. And

According to the present invention, the barcode information collecting means collects the barcode image information described in the object to be recognized, and the barcode information detecting means comprises an x direction or a y direction with respect to the image information collected by the barcode area detecting means. x direction,
By performing a differentiation process consisting of both directions in the y direction, and further performing a threshold process and an edge detection process on the result of the differentiation process,
The image of the barcode may be reconstructed by a barcode recognition unit based on the extracted feature relating to the area where the barcode exists.

[0008] The bar code area detecting means performs a process of detecting an existing point and an edge of the bar code.

The bar code recognition means generates a center line of the area where the bar code exists, and generates code information in a direction orthogonal to the center line.

According to such a barcode recognition device,
Since the feature is extracted by using the differential image of the barcode and the image of the barcode is reconstructed, the amount of information to be processed is smaller than that of the conventional device, the time is short, the stability is high, and the accuracy is high. Barcode recognition can be realized.

[0011]

Embodiments of the present invention will be described below in detail.

FIG. 1 is a block diagram showing a barcode recognition and location according to an embodiment of the present invention.

The bar code recognition unit comprises a camera 2 for picking up an image of a recognition target 1 on which a bar code is written, and a photoelectric conversion unit for converting a bar code image operated by the camera 2 into an electric digital signal. 3, a bar code area detecting section 4 based on a differential image for generating a differential image in the vertical and horizontal directions by differentiating the electrical digital signal from the photoelectric conversion section 3, and a bar code area detecting section 4. From the generated differential image, the change point of the density is captured, and the presence of the barcode is detected by detecting the presence of the barcode with the pair of the change in the positive direction and the change in the negative direction. A bar code recognizing unit 5 is provided based on a differential image that recognizes code information from a direction perpendicular to the detected direction and sends the code information to a recognition result processing unit 6 including a display unit or a printer.

In the bar code area detecting section 4,
The bar code area is obtained by a known method of calculating the circumscribed rectangular coordinates by connecting the blocks of the bar code area from the generated differential image based on the feature that the blocks are continuous at a constant cycle in the bar code area. To be transmitted.

Next, with reference to FIG. 2, the configuration and processing of the bar code area detecting section 4 in the differential image and the bar code recognizing section 5 based on the differential image in FIG. 1 will be described in detail.

An original image 11 shown in FIG. 2 is an example showing a change in density at a position along the x-axis of a barcode described in the x-direction. In the case of the original image 11, it means that two bar codes exist.

In the bar code area detecting section 4, the x direction differentiating section 12 performs x direction differentiation of the bar code developed (described) in the x direction. In this case, the x-direction differentiation processing uses a Sobel filter operator well known in the art.

As a result of such x-direction differentiation, the original image 11
The change in density at each position is converted into the magnitude of the change at each position.

In the bar code area detecting section 4, when the original image 11 is developed in the y direction, x
A y-direction differentiator 13 having the same configuration as the direction differentiator 12 is operated.

If the developing direction of the original image 11 is unknown, x
By operating both the direction differentiating unit 12 and the y-direction differentiating unit 13, either one is selected according to the recognition result.

In the barcode area detecting section 4, the result of each direction differentiation of the x-direction differentiating section 12 and the y-direction differentiating section 13 is output to a positive threshold processing section 14, a negative threshold processing section 15 or an edge direction detecting section 16. Sent to

The positive threshold value processing unit 14 performs threshold processing on only the positive component of the result of the x-direction differentiation to thereby determine the existence point of the barcode on the x-axis (the point α1 on the x-axis in FIG. 2). ) Is calculated.

The negative direction threshold processing unit 15 performs threshold processing by using only the negative component of the result of the x-direction differentiation, so that the existing point on the x-axis of the barcode (on the x-axis in FIG. 2). The point α2) is calculated.

Further, the edge direction detection section 16 detects a positive / negative pair of edges as a result of the x-direction differentiation, and determines the center of the positive / negative pair of edges as an existing point on the x-axis of the barcode (see FIG. 2). x) on the x-axis.

The bar code recognizing unit 5 includes a center line detecting unit 17 based on a period and a bar code information decoder 18.

The results of the above-described positive direction threshold processing, negative direction threshold processing, and edge direction detection processing are as follows:
Any or all of the bar code presence information on the x-axis is sent to the center line detection unit 7 based on the period. The center line detector 7 detects each of the existing points (points α1, α2
, Β), the existence thereof can be confirmed, and this is defined as the center line L.

The bar code information decoder 18 has bar code information in the vertical direction of each existing point (points α1, α2, β) when viewed from the center line L, and is located at a relative position from the center line L. The bar code information is decoded, the processing result is sent to the recognition result processing unit 6, and displayed on the display unit or printed out by the printer.

Next, with reference to FIGS. 3 to 6, the processing in the barcode area detecting section 4 and the barcode recognizing section 5 will be described more specifically.

FIG. 4 shows a multi-valued image of a bar code. In the present embodiment, the differential image shown in FIG. 5 is used because the differential image is uniformly used. The differential image shown in FIG. 5 indicates the presence of information necessary for recognition in the differential image, not the differential image itself. That is, the information at the position shown in black in FIG. 5 is necessary for recognition, and the information at the position shown in white is unnecessary.

Further, since the information at the black position and the information at the white position have continuity as long as the barcode is recognized, the differential image shown in FIG.
Can be significantly compressed. The barcode is recognized from the compressed differential image.

The image shown in FIG. 6 shows differential features obtained by developing a part of the barcode recognition candidate area. In this case, the information at the white position in the differential image shown in FIG. 5 is the same, and only the information at the black position is developed.

The writing direction of the barcode can be estimated from the circumscribed rectangle information of the recognition candidate area.
Detects barcode changes in a fixed cycle.

The outline of the cross section of the image shown in FIG. 6 indicates its existence line. As seen from the existence line of the bar code, there is bar code information in the vertical direction of each existing point. Therefore, from the existing points (points α1, α2, β), the change point of the code information and the edge in the differentiation are calculated. Detected, and the distance between the two points is used as barcode information.

In the recognition of the differential feature in FIG. 6, a bar code is formed with a positive / negative pair. Edge direction detection processing by the edge direction detection unit 16 based on the multi-valued image (multi-valued differential image) shown in FIG. 4 (step S1) (step S2)
Thus, the positive and negative directions and the size of the edge are generated.
In recognition method 1 (step S3), the presence of a barcode is recognized using a positive / negative pair detected by the edge direction detection processing.

On the other hand, the positive / negative threshold processing (step S
In 4), a threshold is determined from the peak of the differentiation and binarized separately for positive and negative to generate a positive differential binary image (step S5) or a negative differential binary image (step S6).

In recognition method 2 (step S7), the presence of a barcode is recognized by using one or both of the positive and negative differential binary images.

Alternatively, in the recognition method 2 (step S7), the existence of the barcode may be recognized independently of the sign, and the recognition result may be edited to recognize the presence of the barcode.

[0038]

According to the present invention, the recognition based on the differential image is performed at a higher speed than the conventional recognition using the multi-valued image, and more accurate than the recognition using the binary image. Can recognize the barcode. That is, the barcode can be recognized at high speed and with high accuracy by using the recognition target image as the differential image.

[Brief description of the drawings]

FIG. 1 is a block diagram illustrating a configuration of a barcode recognition device according to an embodiment of the present invention.

FIG. 2 is an explanatory diagram showing a configuration and a processing state of a barcode area detection unit and a barcode recognition unit in the barcode recognition device of the present embodiment.

FIG. 3 is a flowchart illustrating a flow of a barcode recognition process in the barcode recognition device of the present embodiment.

FIG. 4 is a diagram illustrating an example of a multivalued image applied to the present embodiment.

FIG. 5 is a schematic diagram illustrating an example of a differential image applied to the present embodiment.

FIG. 6 is a schematic diagram illustrating a feature extraction image of a barcode image differentiation process applied to the present embodiment.

[Explanation of symbols]

 REFERENCE SIGNS LIST 1 recognition target 2 camera 3 photoelectric conversion unit 4 barcode area detection unit 5 barcode recognition unit 6 recognition result processing unit 12 x-direction differentiation unit 13 y-direction differentiation unit 14 positive threshold processing unit 15 negative threshold processing unit 16 edge Detector 17 Center line detector 18 Decoder

Continued on the front page (72) Inventor Yoshikatsu Nakamura 70 Toshiba Intelligent Technology Co., Ltd.

Claims (9)

[Claims] 1. A bar code information collecting means for collecting image information of a bar code described on an object to be recognized, and a bar code existence area including a differentiation process for the image information collected by the bar code information collecting means. Bar code area detecting means for performing a feature extracting process; and bar code recognizing means for reconstructing the bar code image based on the feature relating to the area where the bar code is extracted extracted by the bar code area detecting means. A barcode recognition device characterized by the above-mentioned. 2. A bar code information collecting means for collecting image information of a bar code described on an object to be recognized, and an x direction or a y direction for the image information collected by the bar code information collecting means. a barcode region detecting unit that performs a differentiation process including both of the y-direction, a threshold process, and an edge detection process on the result of the differentiation process, and performs a feature extraction process on an area where the barcode exists; A barcode recognition device comprising: a barcode recognition unit configured to reconstruct an image of the barcode based on the extracted characteristics related to the area where the barcode exists. 3. A bar code information collecting means for collecting image information of a bar code described on an object to be recognized, and an x direction or a y direction for the image information collected by the bar code information collecting means. a barcode area detecting means for performing a differential processing consisting of both of the y-direction, a threshold processing and an edge detecting processing for the result of the differential processing, and performing a characteristic extracting processing relating to the area where the barcode exists; Based on the feature relating to the area where the barcode is extracted, the generation of the center line of the area where the barcode is present and the generation of code information in a direction orthogonal to the center line are performed to reconstruct the image of the barcode. A bar code recognition device, comprising: 4. A bar code information collecting means comprising a camera and a photoelectric conversion unit for collecting image information of a bar code described on an object to be recognized, and an x or y direction for the image information collected by the bar code information collecting means. A differentiating means for performing a differentiating process in the x direction or both the x direction and the y direction, a threshold value processing unit for performing a threshold value process on the differential processing result of the differentiating device, and an edge detecting process for the differentiating result of the differentiating device A bar code area detecting means for performing feature extraction processing on the bar code existence point and the edge; and an edge direction detecting section, based on the bar code existence point and the edge related feature extracted by the bar code area detecting means. In addition, the generation of the center line of the area where the bar code is present and the generation of code information in a direction orthogonal to the center line are performed. Barcode recognition apparatus characterized by comprising: a bar code recognition unit for performing image reconstruction of keycode, the. 5. A bar code information collecting means comprising a camera and a photoelectric conversion unit for collecting image information of a bar code described on an object to be recognized, and an x or y direction for the image information collected by the bar code information collecting means. A differential unit for performing a differential process in the x-direction or the y-direction, and a threshold processing unit for performing a threshold process on a result of the differential process performed by the differentiator, and extracting a feature for detecting an existing point of the barcode. A bar code area detecting means for performing processing, and generating a center line of the bar code existing area based on the feature related to the existing point of the bar code extracted by the bar code area detecting means, and generating a center line in a direction orthogonal to the center line. Barcode recognition means for generating code information and reconstructing the image of the barcode. apparatus. 6. The threshold processing unit performs a threshold extraction process in either the positive direction or the negative direction or both of the differential processing results of the differentiating means, and performs a feature extraction process of detecting an existing point of the barcode. The barcode recognition device according to claim 4 or 5, wherein 7. A bar code information collecting means comprising a camera and a photoelectric conversion unit for collecting image information of a bar code described on an object to be recognized, and an x or y direction for the image information collected by the bar code information collecting means. A differentiating means for performing a differentiation process in the x direction or both the x direction and the y direction; and an edge direction detection unit for performing an edge detection process on a result of the differentiation process performed by the differentiating device, and a feature extraction process relating to an edge of the barcode. A bar code area detecting means for performing the generation of a center line of an area where the bar code exists and the generation of a center line of code information in a direction orthogonal to the center line based on the feature relating to the edge of the bar code extracted by the edge direction detecting unit. Barcode recognition means for generating and reconstructing the image of the barcode. Recognition device. 8. A bar code information collecting means comprising a camera and a photoelectric conversion unit for collecting image information of a bar code described on an object to be recognized, and an x direction or y for the image information collected by the bar code information collecting means. A differentiating means comprising an x-direction differentiating unit and a y-direction differentiating unit for performing a differentiation process consisting of directions or both of the x-direction and the y-direction, and performing a threshold process in a positive direction and a negative direction on a result of the differentiation process by the differentiating unit A threshold processing unit including a positive threshold processing unit and a negative threshold processing unit; and an edge direction detection unit that performs positive and negative edge detection processing on the differentiation processing result of the differentiating means. Bar code area detection means for extracting a code existence point and performing a feature extraction process on the bar code edge by a pair of positive and negative edge detection processes; A center line detecting unit that generates a center line of the bar code existing area based on the feature relating to the existing point and edge of the bar code extracted by the bar code area detecting unit; A bar code recognizing device, comprising: a decoder for generating code information; and bar code recognizing means for reconstructing an image of the bar code using the center line and the code information. 9. A barcode information collecting means comprising a camera and a photoelectric conversion unit for collecting image information of a barcode described on an object to be recognized, and an x-direction or a y-direction for the image information collected by the barcode information collecting means. A differentiating means comprising an x-direction differentiating unit and a y-direction differentiating unit for performing a differentiation process consisting of directions or both of the x-direction and the y-direction, and performing a threshold process in a positive direction and a negative direction on a result of the differentiation process by the differentiating unit A threshold processing unit including a positive threshold processing unit and a negative threshold processing unit; and an edge direction detection unit that performs positive and negative edge detection processing on the differentiation processing result of the differentiating means. Bar code area detection means for extracting a code existence point and performing a feature extraction process on the bar code edge by a pair of positive and negative edge detection processes; A center line detecting unit that generates a center line of the bar code existing area based on the feature relating to the existing point and edge of the bar code extracted by the bar code area detecting unit; A barcode recognizing means for reconstructing an image of the barcode based on the center line and the code information; and outputting barcode information reconstructed by the barcode recognizing means. And a recognition result processing means for processing.