A compact underwater stereo vision system for measuring fish

In the context of aquaculture, fish length is a key parameter to assess fish stocks, essential for feeding regime and contributes for decisions at several production levels. However, taking length measurements is a cumbersome task that, when applied to living individuals, can induce considerable stress, increasing the risk of damage or hindering their growth. Computer vision is one of the most used non-contact tools to overcome this issue, being fast, consistent and repeatable. However, its use in aquatic environments is limited by the high cost, the difficulty of calibrating the system in underwater conditions and the complexity of implementation. This paper proposes a low-cost easy-to-use vision system that can take measurements on live fish in aquatic conditions, without the need for a special calibration or a demanding in-water calibration service. The present work implemented a compact stereo vision system and developed a method that estimates the correct length of fish, based on the variation of the angle of incidence of the light rays in the water. Given some structural conditions such as a short baseline, the system is able to measure fish with an error of less than 1%. The short baseline allows to have a compact system and reduces the effect of water refraction on the 3D reconstruction. A set of experiments were performed with real fish, working robustly for a set of orientations of the fish (even when the caudal fin and snout are on different distances to the cameras).