Bill rapid positioning method and system based on color line segments

Abstract

The invention relates to a bill frame rapid positioning method based on color line segment detection and a system thereof. The method comprises the steps of inputting a bill image, scaling the image in proportion, detecting color line segments, merging the line segments, screening the line segments, building a maximum quadrangle and outputting a vertex. The method has the advantages that the colorbills can be recognized, the method is not affected by gray scale conversion, the vertexes of the bills are accurately and rapidly positioned, and the effective bill areas are cut out.

Description

technical field

[0001] The invention relates to a bill rapid positioning method and system based on colored line segments, belonging to the field of image recognition. Background technique

[0002] Through image processing technology, the text and numbers of bills are converted into electronic version data, and the collection of data information is quickly realized through computer programs. In the traditional financial process, there are a large number of bills that need to be entered manually, and the use of image recognition technology greatly improves the efficiency and accuracy of data collection. With the development of image processing and OCR character recognition technology, the electronic financial process has greatly reduced the work intensity of financial bill reimbursement.

[0003] In automatic bill recognition, business card recognition and other scenarios, the image obtained by using the mobile terminal to take pictures is not an accurate bill, and even the ...

Examples

Embodiment 1

[0064] Such as figure 1 As shown, the rapid positioning method of bill frame based on color line segment detection includes the following steps:

[0065] S1: Obtain a frame of image containing bills, such as figure 2 The original image is shown in color.

[0066] S2: Perform proportional reduction processing on the one frame of the image containing the receipt by using a linear interpolation method to obtain a reduced image. The proportional value lv=300 / min(w, h) used by the linear interpolation method, wherein, w, h are respectively the width and length of the image containing the bill in the frame, and the width and height of the reduced image are respectively w 2 =w×lv,h 2 =h×lv.

[0067] S3: Through the color LSD line segment detection algorithm, identify all the line segments in the reduced image, the line segments are expressed as L={x1, y1, x2, y2}, where (x1, y1) and (x2, y2) are The two endpoints of the line segment L, and all the line segments form a line segm...

Embodiment 2

[0092] The rapid positioning system for bill borders based on color line segment detection includes a memory and a processor, the memory stores instructions, and the instructions are suitable for being loaded by the processor and performing the following steps:

[0093] Get a frame containing the image of the ticket.

[0094] The frame of the image containing the bill is reduced proportionally by a linear interpolation method to obtain a reduced image. The proportional value lv=300 / min(w, h) used by the linear interpolation method, wherein, w, h are respectively the width and length of the image containing the bill in the frame, and the width and height of the reduced image are respectively w 2 =w×lv,h 2 =h×lv.

[0095] Use the color LSD detection algorithm to find all the line segments in the reduced image to form a line segment set L_set, the line segment is expressed as L={x1, y1, x2, y2}, where (x1, y1) and (x2, y2) is the two endpoints of the line segment L, wherein, t...

technical field

[0001] The invention relates to a bill rapid positioning method and system based on colored line segments, belonging to the field of image recognition. Background technique

[0002] Through image processing technology, the text and numbers of bills are converted into electronic version data, and the collection of data information is quickly realized through computer programs. In the traditional financial process, there are a large number of bills that need to be entered manually, and the use of image recognition technology greatly improves the efficiency and accuracy of data collection. With the development of image processing and OCR character recognition technology, the electronic financial process has greatly reduced the work intensity of financial bill reimbursement.

[0003] In automatic bill recognition, business card recognition and other scenarios, the image obtained by using the mobile terminal to take pictures is not an accurate bill, and even the ...

Examples

Embodiment 1

[0064] Such as figure 1 As shown, the rapid positioning method of bill frame based on color line segment detection includes the following steps:

[0065] S1: Obtain a frame of image containing bills, such as figure 2 The original image is shown in color.

[0066] S2: Perform proportional reduction processing on the one frame of the image containing the receipt by using a linear interpolation method to obtain a reduced image. The proportional value lv=300 / min(w, h) used by the linear interpolation method, wherein, w, h are respectively the width and length of the image containing the bill in the frame, and the width and height of the reduced image are respectively w 2 =w×lv,h 2 =h×lv.

[0067] S3: Through the color LSD line segment detection algorithm, identify all the line segments in the reduced image, the line segments are expressed as L={x1, y1, x2, y2}, where (x1, y1) and (x2, y2) are The two endpoints of the line segment L, and all the line segments form a line segm...

Embodiment 2

[0092] The rapid positioning system for bill borders based on color line segment detection includes a memory and a processor, the memory stores instructions, and the instructions are suitable for being loaded by the processor and performing the following steps:

[0093] Get a frame containing the image of the ticket.

[0094] The frame of the image containing the bill is reduced proportionally by a linear interpolation method to obtain a reduced image. The proportional value lv=300 / min(w, h) used by the linear interpolation method, wherein, w, h are respectively the width and length of the image containing the bill in the frame, and the width and height of the reduced image are respectively w 2 =w×lv,h 2 =h×lv.

[0095] Use the color LSD detection algorithm to find all the line segments in the reduced image to form a line segment set L_set, the line segment is expressed as L={x1, y1, x2, y2}, where (x1, y1) and (x2, y2) is the two endpoints of the line segment L, wherein, t...

Images