Cow face and cow face key point detection method based on Mask-RCNN

Abstract

The invention discloses a cow face and cow face key point detection method based on Mask-RCNN. The blank of cow face and cow face key point synchronous detection is filled. The cow face and cow face key point detection method based on Mask-RCNN comprises the following steps: 1) collecting pictures with cow faces, and marking the cow faces and cow face key point data in each picture: (1) collectingthe pictures with the cow faces from a domestic network, and selecting the pictures with the cow faces from the collected pictures: a, selecting a video related to cow raising from domestic agricultural programs, intercepting one picture every five frames, and selecting the pictures with the cow faces from the pictures with the cow faces; b, collecting a video of the cow on site by using a camera, intercepting a picture every five frames, and selecting a picture with a cow face from the pictures; (2) marking the position of each cow face in the picture; 2) designing a cow face and cow face key point detection network structure, and 3) training a cow face and cow face key point detection network by using the marked data to finally generate a cow face and cow face key point detector.

Description

technical field [0001] The invention relates to a detection method belonging to the technical field of digital image processing, more specifically, the invention relates to a method for detecting a cow face and key points of a cow face based on Mask-RCNN. Background technique [0002] Animal identification can facilitate production management, control disease outbreaks and establish ownership, and also meet the needs of today's consumers for traceability and import and export trade requirements. [0003] Modern animal identification technologies include mechanical (such as branding, tattoos, etc.), electronic (such as ear tags, RFID rumen ceramic tags, implantable ID chips, etc.), biometrics (such as nose prints, DNA atlas, iris recognition, etc.). Both mechanical and electronic animal identification methods have the disadvantages of difficult operation, low animal welfare, high cost, high loss rate, and low reuse rate. [0004] The cattle face recognition technology is a ...

AI Technical Summary

Problems solved by technology

The Chinese patent announcement (cloth) number is CN107292298A, and the announcement (cloth) date is August 9, 2017. There is no mention of cow face in the case of the invention titled "Cow Face Recognition Method Based on Convolutional Neural Network and Classifier Model" The problem of detection is just to adjust the angle of view of the camera and manually select pictures with only one cow face in the image to form a training set and a test set. It is conceivable that this method is not very versatile.

[0009] 1. There is no cow face detection method, but the general method of target detection is used, which leads to low accuracy and low professionalism of cow face detection

[0010] 2. The number of key points of the cow face is not analyzed, and there is a lack of a more accurate detection method for the key points of the cow face

[0011] 3. The data is not uniform, and the noise of the image is not wide, resulting in poor generalization of the training model

[0012] 4. The multitasking is not strong. The cow face detection and the key point detection of cow face are independent portals. They are all carried out step by step, and the two tasks are not completed in a unified manner.

Images