Fish fall sick, SingTown Technology OpenMV knows first.

In aquaculture and ornamental fish management, common diseases such as fish skin diseases, fin and tail rot, and water mold disease, if not detected in time, will severely affect the health and survival rate of fish populations. Traditional detection mainly relies on manual visual observation, which is not only inefficient but also easily interfered with by turbid water, changes in light, and fish movement, making it difficult to accurately and timely determine pathological conditions.

Therefore, aquaculture enterprises adopt SingTown Technology's OpenMV intelligent camera to establish a machine vision-based fish health monitoring system for real-time identification of abnormalities on fish surfaces. This system is installed on the outer glass walls of fish tanks or aquaculture ponds, capable of monitoring fish activities within a range of 1 meter, and automatically identifying and detecting lesions in small goldfish approximately 5 centimeters in length.

The system's objective is to determine whether lesions such as white spots, white fuzz, water mold, or fin rot appear on the fish's body surface through image recognition, thereby providing early health warnings and assisting breeders in timely intervention.

Aquaculture enterprises utilize the SingTown intelligent camera system to construct a fish health identification framework, integrating AI image recognition, color segmentation, texture analysis, and feature comparison algorithms, enabling precise detection of fish body lesions underwater.

The system captures real-time images of the aquarium using a high-definition camera. The edge AI employs a color recognition algorithm to extract the fish body area and filters out background reflections. By analyzing the brightness distribution and texture continuity on the fish surface, the algorithm identifies abnormal reflective areas such as white spots and water mold. Additionally, the algorithm can assess the presence of fin rot, fin decay, or fungal attachments by combining morphological features, such as irregular fin edges, localized whitening, or blurring.

To adapt to the movement of fish in water and fluctuations in light, the system incorporates dynamic threshold adjustment and multi-frame fusion algorithms, providing stable judgment for short-term continuous images and effectively reducing false alarms. In testing, with the recognition distance controlled within 1 meter, the system can simultaneously identify multiple individuals, achieving an accuracy rate of over 92% in detecting water mold disease in small fish such as goldfish. Detection results can be output in real-time to the host computer or APP, sending alerts such as suspected lesions via the network.

SingTown OpenMV Smart Camera is easy to operate, low in power consumption, and strong in real-time detection. It is suitable for deployment on edge devices in aquaculture, pet aquariums, and university research laboratories, helping users achieve timely, visual, and intelligent monitoring of fish health.