Traditional shrimp farms rely on the farmers' experience to determine the shrimp's appetite, and thus, the amount of feed to provide the shrimp during each feeding time. Usually, a net is required to hoist the bottom-dwelling shrimp in order to observe their appetite, however, this method is quite difficult to perform and does not provide immediate results, which leads to farmers often providing inaccurate amount of shrimp feed. This is detrimental to the farm as excess feed leads to wastage and reduces the water quality, whereas insufficient feed will affect the shrimp's growth. By capitalizing on the rapid development of technology in the fields of Internet of Things. artificial intelligence, and big data, modern aquaculture industries have been able to not only increase their production quantity and quality, but also overcome various problems encountered by traditional aquaculture industries. However, the complex and challenging nature of underwater monitoring has, thus far, prevented the aquaculture and fishing industry—especially the shrimp farming industry—from enjoying the benefits of intelligent farm supervision and management.
In this study, a system to monitor and determine shrimp appetite is developed (Figure 1). A small amount of feed is first cast on a monitoring platform, then, an innovative underwater image capture system is used to acquire images of shrimp eating the feed. The images are subsequently transmitted to a cloud server, where artificial intelligence is used to analyze the shrimp's appetite in real-time, and evaluate the amount of feed to provide as well as the shrimp's growth trend. Finally, the evaluation results are used to automatically feed the whole shrimp population in the farm. In addition, all the data obtained from the aforementioned processes are also transmitted to the user's (i.e. farmers) dashboard, providing them with real-time aquaculture data, allowing them to work remotely.
The artificial intelligence technology of this system is the core of the union between intelligence and aquaculture. Using said technology to analyze shrimp appetite and decide on the amount of provided feed reduces the time and labor requirements of manual observation in traditional shrimp farms and their associated human errors, improves the farm's productivity, and even augment shrimp growth. Moreover, the integration of an innovative two-stage shrimp feeder allows the system to perform the feeding automatically, with minimal human intervention.
The proposed system introduces artificial intelligence as a method for shrimp aquaculture management, is able to solve the various problems of traditional shrimp aquaculture, and brings a lot of positive benefits. With respect to the industry, the system can directly improve aquaculture efficiency while reducing costs, thus, increasing the output value of the shrimp aquaculture industry. In terms of technology, the innovative design of the underwater camera can be applied in various aquaculture environments, and the development of artificial intelligence models also provides a reference for future applications in the industry. With regards to the environment, intelligent feeding strategies can reduce feed waste as well as the possibility of waste water polluting the environment. As to society, the development of smart shrimp farming systems can drive the advancement of smart agricultural policies. Finally, this system can be applied to aquaculture fields of various scales, and is able to analyze different phases according to the requirements of the users, providing a selection of flexible business models.
Comments