Automatic Penaeus Monodon Larvae Counting via Equal Keypoint Regression with Smartphones

Simple Summary To enhance farming efficiency, most farmers prefer to purchase Penaeus larvae as opposed to hatching them themselves. However, counting small and highly congested larvae during transactions is a challenging task to accomplish manually. We intend to improve counting precision and decrease human labor costs. In this work, an equal keypoint regression method is proposed to address these challenges. We employed five different types of smartphones to capture thousands of high-resolution images under various challenging environmental conditions. Then, we selected 1420 images to build a high-resolution dataset. In addition, this high-resolution dataset included general point annotations for use. Following training with this dataset, we obtained a model that we tested with a real Penaeus monodon larvae dataset. The results showed that the average model accuracy for the 720 images with seven density groups in the test dataset was 93.79%. Ultimately, our trained model demonstrated greater efficiency than the classical density map algorithm. Today, large-scale Penaeus monodon farms no longer incubate eggs but instead purchase larvae from large-scale hatcheries for rearing. The accurate counting of tens of thousands of larvae in these transactions is a challenging task due to the small size of the larvae and the highly congested scenes. To address this issue, we present the Penaeus Larvae Counting Strategy (PLCS), a simple and efficient method for counting Penaeus monodon larvae that only requires a smartphone to capture images without the need for any additional equipment. Our approach treats two different types of keypoints as equip keypoints based on keypoint regression to determine the number of shrimp larvae in the image. We constructed a high-resolution image dataset named Penaeus_1k using images captured by five smartphones. This dataset contains 1420 images of Penaeus monodon larvae and includes general annotations for three keypoints, making it suitable for density map counting, keypoint regression, and other methods. The effectiveness of the proposed method was evaluated on a real Penaeus monodon larvae dataset. The average accuracy of 720 images with seven different density groups in the test dataset was 93.79%, outperforming the classical density map algorithm and demonstrating the efficacy of the PLCS.