Livestock Weight Detection Using Computer Vision-Based Smartphone App for Accurate Service Delivery

Shafiq-ul Islam of ACME AI in Bangladesh will produce a smartphone-based system that uses computer vision and machine learning to accurately estimate the weight of cows and goats to help smallholder livestock farmers in rural Bangladesh maximize productivity and profits. Accurately determining livestock weight is challenging for these farmers but critical for determining the right amounts of food and medicines. They will develop a machine learning model and mobile application that uses the smartphone’s camera to process distance, height, and depth information and calculate the weight of the animal to within >90% accuracy. They will test three different business cases, including combining the computer vision-based weighing system with products and service providers, and evaluate the impact on food and medicine purchases, and animal growth and quality, which are directly linked with income.