Live body weight monitoring method for animal husbandry based on Internet of Things

Abstract

The invention discloses a live body weight monitoring method for animal husbandry and breeding based on the Internet of Things. Data is collected by a camera and a sensor. After the data collection is completed, the data is transmitted to a single-board computer, and the data is integrated by the single-board computer, and the integrated data is transmitted to The local terminal C / S carries out the next step of processing; on the local terminal C / S, the collected RGB image is firstly recognized for living body, combined with the depth point cloud data of the depth camera, a 3D point cloud image is derived; at the same time, the RGB, The 3D point cloud image, as well as the individual information and environmental variables of the living body collected by the sensor are uploaded to the cloud server; the cloud server cleans and denoises the 3D point cloud image according to the uploaded individual information and environmental variables, and restores the volume after reaching the standard. After the living body volume is calculated, the mass is calculated; the result of the living body mass is returned to the C / S end. If there is any objection to the quality error, manual corrections can be made, and the corrections can be recorded and uploaded to the cloud server.

Description

technical field [0001] The present invention relates to the field of agricultural internet of things, in particular to a method for monitoring live body weight in animal husbandry based on agricultural internet of things. Background technique [0002] In 2015, China's GDP was 67.67 trillion yuan, of which agriculture was 9 trillion yuan, and animal husbandry was more than 3 trillion yuan. In 2015, there were nearly 7,000 feed companies in China, 500 of which had an annual output value of more than 500 million yuan; nearly 2,700 designated pig slaughtering companies above designated size; more than 20,000 pig farms with an annual slaughter of more than 5,000 heads. The period from 2010 to 2016 is a period of rapid integration of the agriculture and animal husbandry industry. The number of feed companies has decreased from 11,000 to 6,000; the number of pig slaughtering companies has decreased from 40,000 to 17,000; the scale of pig breeding has increased from 30% to 60%; 2020...

AI Technical Summary

Problems solved by technology

However, there are indeed a series of problems in the agricultural and animal husbandry industry data: A. Data islands

The current breeding technology is mainly human-intervened floor scale weighing. This method can easily cause stress reactions in pigs. Pigs are prone to stress reactions such as anorexia, vomiting, and weight loss. Therefore, farms cannot carry out frequent weighing. Impossible for precision farming

[0004] At present, the following technologies are disclosed, but the weight data obtained by these methods are easily affected by the activities of pigs, resulting in inaccurate measurement, unstable results, and limited data, which is not conducive to big data monitoring

Images