The Design And Implementation Of Harmful Algal Bloom Intelligent Video Surveillance System

With the expansion of population,resource shortages and environmental degradation,the protection of marine living resources is particularly important.As a big maritime country,China's development of the marine economy has become one of the current national strategies.Algae occupies a special position in marine living resources.It can produce a large amount of organic matter through photosynthesis,provide sufficient food for marine animals,and release a large amount of oxygen in photosynthesis,providing essential for marine animals and even land-based organisms.However,due to special environmental conditions,certain algae organisms are over-reproduced and highly aggregated,that is,"Harmful Algal Blooms(HABs)",which cause marine algae pollution.In the past decade,China's marine harmful algal blooms have become increasingly serious.From a small number of scattered sea areas to a piece of sea area,some important aquaculture and tourism industries are seriously threatened and lost.In order to realize marine environment monitoring and harmful algal bloom warning,this paper designs and implements a harmful algal bloom intelligent video monitoring system.On the one hand,the system realizes real-time monitoring of the on-site environment,and the live broadcast of video facilitates the real-time viewing of the situation by researchers.On the other hand,the system realizes the automatic identification and classification of a variety of harmful algae,and thus completes the intelligent classification and storage of video,further improving the research efficiency of marine ecological researchers.In addition,the system service interface is based on the B/S mode,and the excellent cross-platform features greatly increase the application range and application scenarios of the system.The harmful algal bloom intelligent video surveillance system architecture designed in this paper is divided into three parts from the bottom up,namely video acquisition layer,edge computing layer and ocean cloud platform layer.First,the video acquisition layer is based on the ARM architecture,realizing the real-time collection and transmission of marine algae data,providing data support for the subsequent processing of the system.Secondly,in order to provide low-latency video live broadcast service,this paper uses the edge computing layer to be responsible for the video transfer.In addition,the edge layer realizes automatic identification and storage of marine algae based on neural network,which alleviates the calculation pressure and storage pressure on the cloud platform side.Finally,the cloud platform layer implements high-performance,persistent storage of monitoring data,and at the same time,implements a unified service interface for the system to provide high-performance video services for researchers.The three-tier architecture of this paper adopts the producer-consumer model,which realizes the decoupling between producers and consumers and enhances the scalability of the system.At the same time,the overall performance of the system is improved by balancing the processing power of the consumers and consumers.This paper has carried out the architecture and technical exploration for the realization of marine ecological monitoring system,which has good social significance and application value.