Mobile system for monitoring body sizes and weights of pigs in multiple pigsties

Abstract

The invention relates to a mobile system for monitoring body sizes and weights of pigs in multiple pigsties. The radio frequency technology, the binocular vision technology and the automatic control technology are adpoted in the system, and body sizes and weights of pigs in the multiple pigsties of a hog house are automatically monitored. The system comprises a pig individual recognition device, a camera moving device and a binocular vision device. By the adoption of the system, body sizes and weights of pigs in the multiple pigsties can be monitored without causing stress to pigs or affecting growth of pigs. The system is high in automation degree, easy to operate, capable of adapting to the structure of the hog house and low in cost.

Description

technical field [0001] The invention relates to the technical field of automated animal husbandry, and more particularly relates to a mobile multi-sty pig body size and weight monitoring system. Background technique [0002] In pig breeding, parameters such as pig body length, body width, body height, hip width, hip height, and weight are important indicators for describing the growth status and reproductive performance of pigs. The growth information of live pigs can be monitored in time, and precise feeding can be carried out according to body weight, so as to predict the date of market slaughter and improve the weight uniformity of market pigs. In traditional pig farming, these growth parameters are generally measured manually. Body measurements and other information are measured with a tape measure or stick, and body weight is measured with a weight cage and scales. These manual measurement methods are time-consuming, labor-intensive, inefficient, and cause great stres...

AI Technical Summary

Problems solved by technology

This method solves the stress problem very well, but the equipment described in this method cannot be moved and can only be installed in a fixed position

Most domestic pigs are raised in small pens. This method can only detect the weight of pigs in one pen, and detect the weight of all pigs in a pig house. It requires multiple sets of equipment, and the cost is high, so it is difficult to apply in practice.

This patent mainly uses three parameters of the pig's back area and pig body height to estimate the body weight. The model parameters are few and the accuracy is not high.

In addition, the calibration steps of the system are complicated, and it is necessary to manually correspond to the spatial coordinates and image coordinates of the calibration board, and bring them into the model for solution. Once the camera moves, it needs to be re-calibrated

This method takes the ground as the origin of the Z axis, and assumes that the ground is horizontal, but the actual ground of the pig house has a certain drainage angle, which causes the detection error of the body size

[0004] Australian AU2010219406 and US patent US20130064432 use image processing technology to automatically detect the pig’s back area and estimate the pig’s weight. This method cannot detect the pig’s body height and body size. The regression model established with one parameter pixel area and pig weight is unreliable and requires a camera stay at a unique height

This method also cannot detect the weight of pigs in multiple pens. If it is erected in the form of aisle, the original pig house needs to be modified, which also increases the cost of the system.

[0005] There are also handheld mobile detection equipment for cattle and sheep abroad, but this type of equipment is only for a single animal, and many of them need to limit the animal and manually select the measurement point, which cannot realize large-scale automatic monitoring

Images