A combination of IoT and cloud application for automatic shrimp counting

The pet market is getting growth rapidly in the world, and the ornamental fish occupy the third in the market ranking, behind dogs and cats. According to the statistics of the Ornamental Fish Association, Taiwan has exported 18 million ornamental shrimps annually since 2010. Almost six-tenths of global ornamental shrimps are from Taiwan. OpenCV (Open Source Computer Vision) provides plenty of machine vision applications and often cooperates with the Raspberry Pi to enhance the use of machine engineering for commercial products. This research is, therefore, mainly designed to apply the machine vision to undertake the counting of shrimps automatically. The steps of image processing for accurately counting shrimps are as follows: (1) read the image graphic, (2) filter and remain the sampling color, (3) threshold the image, (4) contour the shrimps in the image (5) count the number. Concerning the performance and reliability, we process the image using Amazon Web Service (AWS) Lambda function. Experimental results of counting shrimps ( Neocaridina heteropoda var. red ) show that it takes 0.1 s to count 150 shrimps and the precision rate is about 95%.