Click on “Download PDF” for the PDF version or on the title for the HTML version.

If you are not an ASABE member or if your employer has not arranged for access to the full-text, Click here for options.

Abstract.

Autonomous detection cow and its key body parts is of significance to precision livestock farming. The advancement in sensor technology, deep learning and field robotics have paved the way for farm management. In this study, a deep learning network named YOLOv4-CSP was used, to achieve the detection of key parts of dairy cows in complex scenes. In order to verify the effectiveness of the algorithm, a challenging cow dataset consisting of adult and calf with complex environments (e.g., day and night) was constructed for experimental testing. The proposed YOLOv4-CSP based approach achieved a precision of 98.00%, a recall of 99.00%, an F1 score of 99.00%, and an mAP@0.5 of 96.86%. Experimental results demonstrated that the proposed YOLOv4-CSP approach could capture key biometric-related features for cow visual representation, improving the performance of cow detection. Overall, the proposed deep learning-based cow detection approach is favorable for long-term autonomous cow monitoring and management in smart livestock farming.