Development of an IoT-Based Crayfish Breeding Monitoring and Automatic Feeding System with Image Processing.

Crayfish farming is steadily growing in the Philippines, especially among small-scale farmers looking for alternative sources of income. Despite its potential, many farmers still rely on traditional and manual methods to monitor water quality and feed their crayfish. These practices often result in poor water conditions, overfeeding, and lower survival and breeding rates. This study developed a prototype system that uses Internet of Things (IoT) sensors, automated feeding, and image processing to make crayfish farming more efficient and manageable. The system monitors key water parameters temperature, pH, dissolved oxygen, and total dissolved solids and sends email alerts if values go beyond the recommended range. It also includes an automatic feeding system that dispenses food at a set time each day to reduce waste and ensure proper nutrition. To support breeding, a USB camera paired with a YOLOv8 image processing model was used to detect gravid (pregnant) crayfish. The system was tested over two weeks in a small-scale setup in Pampanga. Results showed that the water quality stayed within ideal levels, and the image processing model was able to detect pregnant crayfish with reasonable accuracy. Five aquaculture experts who evaluated the system said it was useful, easy to understand, and applicable for real farm use. While improvements can still be made such as increasing detection accuracy, adding backup power, or offline data storage the results suggest that this kind of system can help small-scale crayfish farmers save time, reduce errors, and improve overall productivity.