Rectangular flag bit detection method based on pixel extraction

A technology of pixel extraction and detection method, applied in the direction of instrument, calculation, character and pattern recognition, etc., can solve the problems of reducing program stability, poor compatibility, high demand for printing and scanning, and improve stability and scalability. Effect

Pending Publication Date: 2022-05-31

南京红松信息技术有限公司

0 Cites 0 Cited by

AI-Extracted Technical Summary

Problems solved by technology

The traditional method has two disadvantages. The first disadvantage is that the mark of the answer sheet needs to be a special color, which requires high printing and scanning, and there is a certain threshold for users, and the compatibility is poor.

The second disadvantage is that the area around the s...

Abstract

The invention relates to a rectangular flag bit detection method based on pixel extraction, and the method comprises the steps: S1, carrying out the image restoration and feature extraction of a collected image, and obtaining an average pooling feature map; s2, performing global flag bit fuzzy search on the averaged pooling feature map obtained in the step S1 by using image features to obtain all flag bits possibly existing in the image; s3, carrying out precise positioning reduction on each zone bit obtained in the image; and S4, returning flag bit information conforming to the standard. The detection method for the rectangular flag bit based on pixel extraction not only meets the requirement of intelligent paper marking, but also can be suitable for different scenes, and the stability and expandability of a program are improved.

Application Domain

Character and pattern recognition

Technology Topic

Computer graphics (images)Feature extraction +3

Image

Examples

Experimental program(1)
Effect test(1)

Example Embodiment

Embodiment: as shown in Figure 1, this detection method based on the rectangle mark position of pixel extraction, specifically comprises the following steps

S1: to the test paper image of collection (can be scanned by scanner, or carry out image restoration and feature by taking pictures

The concrete steps of described step S1 are:

S11: global binarization process to the collected image, then detect the outline of the image, and fill in the hole, the most

feature map; the specific process of the average pooling processing in the step S12 is: the restored image to be processed according to the fixed size

S2: the feature map of the average pooling that obtains in step S1 utilizes image feature to carry out the fuzzy search of global sign position

S21: carry out the detection of the image outline again according to the feature map of the average pooling obtained in step S12, for

Described condition one in described step S21 is: outline needs to satisfy that black pixel proportion in outline is greater than 95%;

PUM

Description & Claims & Application Information

We can also present the details of the Description, Claims and Application information to help users get a comprehensive understanding of the technical details of the patent, such as background art, summary of invention, brief description of drawings, description of embodiments, and other original content. On the other hand, users can also determine the specific scope of protection of the technology through the list of claims; as well as understand the changes in the life cycle of the technology with the presentation of the patent timeline. Login to view more.

Similar technology patents

Control method for upper oil pump frequency converter DCAC of electric sanitation truck

InactiveCN103774595AImprove stability and scalabilityEasy troubleshooting

Owner:BEIJING ZHIXING HONGYUAN AUTOMOBILE CO LTD

Multispectral image compression method based on square matrix transformation

InactiveCN105719322AImprove stability and scalabilityGuaranteed Spectral Accuracy and Colorimetric Accuracy

Owner:SICHUAN YONGLIAN INFORMATION TECH CO LTD

Classification and recommendation of technical efficacy words