Image Processing Based Automatic Identification of Freshness in Fish Gill Tissues

Fish is one of the healthiest foods on the earth. The quality of the fish gets changed by the process involved in preserving and handling from buyer to seller. The degradation in the fish can be detected on the most vulnerable focal tissue like gill. This focal tissue has been automatically segmented from whole fish using image processing based technique. The statistical features were extracted from the segmented gill tissues. This paper presents a nondestructive image processing based technique for extracting the properties of stored fish gill tissues using Wavelet Transform. The imperative selection of the methodology makes the algorithm computationally fast, efficient, and automatic. The strategic selection of the discriminatory features from the fish image makes this a novel approach which is accurate for the detection of freshness in the sample under test. Experimental results show discriminatory variation pattern of statistical wavelet feature versus freshness of the fish. This prominent monotonic dissimilarity in the parameters provides a strategic framework for the identification of freshness in stored fish gill tissues. The computation speed is fast which makes this method efficient for real time application.