Cow Estrus Status Identification and Individual Identification System Based on Voiceprint Data

This paper proposes a cow estrus status identification and individual identity recognition system based on voiceprint data. By introducing deep learning and voiceprint recognition technology to replace the traditional manual monitoring of cattle, we can achieve accurate monitoring of the estrus status of cattle. It can effectively solve the problem of missed and wrong detection of estrus cows in large-scale beef cattle breeding scenarios. The method includes: building a data collection system through collars, pickups and high-definition network cameras, using the data collection system to build a cattle voiceprint database and obtaining a high-quality cattle estrus cry data set, and cascade training Conv-TasNet and EcapaTdnn two voiceprint field depths Learn the model to accurately and effectively identify the estrus state and identity of cattle. This research provides an efficient and accurate solution for cattle estrus identification and individual identification. It is expected to be widely used in the breeding industry, improve breeding efficiency, reduce labor costs, and make a positive contribution to the development of the livestock breeding industry.