Shrimp are widely consumed as one of the most favorable seafood in recent years.  As worldwide shrimp production grows, shrimp quality evaluation becomes a critical task for the seafood and aquaculture industries. Automatic evaluation of shrimp shape is critical to improving shrimp quality and production efficiency. This paper proposes an Evolution COnstructed (ECO) features based method to automatically evaluate shrimp shape completeness. Rather than depends on human expert-designed features or deliberated image processing techniques, the proposed method automatically constructs features that are used by Adaboost model to detect broken shrimp.  Experimental results show that ECO features based method obtains an average of 95.1% classification accuracy with a 0.948 precision rate and a 0.920 recall on our shrimp dataset. Although the experiment was performed on our shrimp dataset to prove feasibility, the proposed method can be easily adapted for other shrimp species.

Graduate Students:

Kirt Lillywhite and Beau Tippetts

1. D. Zhang, K.D. Lillywhite, D.J. Lee, and B.J. Tippetts, “Automatic Shrimp Shape Grading Using Evolution Constructed Features,” Computers and Electronics in Agriculture, vol. 100, p. 116-122, January 2014.