Chinese scientists have developed a lightweight model for beef cattle behavior recognition from quadruped robot video in grassland pastures, improving efficiency of herd feeding and controlling.

The lightweight model MASM YOLO was proposed by agricultural information institute of the chinese academy of agricultural sciences and the relevant research was publish in computer and electronics in agriculture.

James Web Telescope

Accurate rapid identification of cattle behaviors is fundamental to disease diagnosis estrus monitoring calving prediction and health assessment.

MASM Yolo enable precise multi behavior detection under complex conditions suitable for real time execution on board a mobile robot.

MASM YOLO gain rapid recognition of six typical behaviors of beef cattle consisting of feeding, resting, locomotion and licking.

It strikes optimal balance between recognition accuracy and computational efficiency.

The model provide technical help for full scale development of grazing robots.

MASM YOLO enable precise multi behavior detection under complex environment for  real time execution on board a mobile robot.

Space X

By integrating the Multi Scale Focus and Extraction Network the Adaptive decomposition and alignment head and technologies MASM YOLO address key challenges consisting of lighting variations motion blur and occlusions within cattle groups.