Fish counting is automated by combining computer vision with our proprietary, bio-inspired fish simulation and realistic environment rendering
Working with SoftBank Corp. and Nosan Corporation, NeuralX, Inc. published a scientific paper at SIGGRAPH Asia 2021, a top tier international conference in the field of computer graphics and interactive technologies. NeuralX presented the paper at Tokyo International Forum in Tokyo on December 14th, 2021.
(a): Current fish counting method at fish farms. Fish are manually lifted in the air from one fish cage and visually counted when they are released to a new cage. (b) : Our new, automated fish counting system running on captured video at a fish farm. The trained computer vision model is able to count the number of fish with high accuracy (97%). (c): Bio-inspired fish simulation and realistic environment rendering which can synthesize training dataset for computer vision.
Paper
Foids: Bio-Inspired Fish Simulation for Generating Synthetic Datasets
Background Story
Food sustainability is one of the impending global challenges. Due to an increasing human population and environmental changes, action must be taken to develop a sustainable food supply to meet the current and near-future food consumption rate. Fish farming is one of the solutions; however, it is still in its early stages when it comes to the incorporation of technology. Surprisingly, fish counting, sizing, and weighting is a very manual process. Currently, this is accomplished by lifting the fish out of the tank, measuring on scales, and visually counting while they are transferred from one cage to another. This process is time-consuming and damages the fish body, which can unnecessarily stress or kill the fish. Regardless of the risks, this measuring is inevitable because fish farms need to know the information to efficiently feed the appropriate amount of food. Hence, a healthier and more efficient solution to measure is highly in demand.
Abstract
We present a bio-inspired fish simulation platform, which we call "Foids", to generate realistic synthetic datasets for an use in computer vision algorithm training. This is a first-of-its-kind synthetic dataset platform for fish, which generates all the 3D scenes just with a simulation. One of the major challenges in deep learning based computer vision is the preparation of the annotated dataset. It is already hard to collect a good quality video dataset with enough variations; moreover, it is a painful process to annotate a sufficiently large video dataset frame by frame. This is especially true when it comes to a fish dataset because it is difficult to set up a camera underwater and the number of fish (target objects) in the scene can range up to 30,000 in a fish cage on a fish farm. All of these fish need to be annotated with labels such as a bounding box or silhouette, which can take hours to complete manually, even for only a few minutes of video. We solve this challenge by introducing a realistic synthetic dataset generation platform that incorporates details of biology and ecology studied in the aquaculture field. Because it is a simulated scene, it is easy to generate the scene data with annotation labels from the 3D mesh geometry data and transformation matrix. To this end, we develop an automated fish counting system utilizing the part of synthetic dataset that shows comparable counting accuracy to human eyes, which reduces the time compared to the manual process, and reduces physical injuries sustained by the fish.
Please find more details in our paper.
About SIGGRAPH Asia
SIGGRAPH (Special Interest Group on Computer Graphics and Interactive Techniques) is an annual conference on computer graphics (CG) organized by the ACM SIGGRAPH, starting in 1974. The main conference is held in North America; SIGGRAPH Asia, a second conference held annually, has been held since 2008 in countries throughout Asia.
【Contact】
NeuralX, Inc. PR team E-mail：info@neuralx.ai