Abstract

Fish is an important source of protein in most countries in the world. The need to know the reproduction and population of fish is crucial for optimum exploitation of fish resources in maintaining the requirement of mankind in the future. In fisheries research, the length of a fish is the main parameter needed to identify fish reproduction, recruitment, growth and mortality. Current method used to acquire these length samples could be problematic as it is manually done; the fish need to be purchased in large quantities and then measuring one by one is time consuming and imprecise. The manual process may lead to overflowing cost. The fish length from digital images (FileDI) framework attempts to avoid this problem using a combination of optical theory and image processing techniques that automatically measures the length of the fish. It reduces cost, faster than previous method and yields accurate length measurement. Preliminary test has shown that the confident level of the FiLeDI framework accuracy is as high as 95% for fish length measurement.

Key words: Fish length, image processing, optical theory.