Identification of Fish Freshness Through Gill Color Using CNN and L*A*B Image Segmentation

In the Philippines, fish makes up a considerable amount of the food supply from the fishing industry and aquaculture. Of these, species such as milkfish and tilapia are common choices for consumption. Fish can be rotten over a period when exposed to air, which can be observed through the different parts of the fish such as the eyes, scales, and gills. The main goal of the study was to produce a system to identify the freshness of a fish gill. This was conducted by segmenting the image using the value of the A channel in the L*A*B color space. The species of fish included milkfish (bangus), tilapia, and catfish species. The accuracy of the system reached up to 90%.