Software Design Of Intelligent Biometric Towed Underwater Microscope

Along with the ocean is more and more thorough in being understood and developed, it becomes more and more necessary and urgent for us to monitor the marine environment. Because of the relation to the marine nitrogen cycle, the number and distribution of marine nitrogen fixation creature is a crucial marine environment data. Due to its ability to identify the marine nitrogen fixation creature in real time and accuracy, and offer various statistics such as the number and distribution, the intelligent biometric towed underwater microscope software system researched and developed by this thesis has high value of application.The software system in this thesis is designed based on the TI Da Vinci platform and PC, and on the basis of its function, it has been divided into lower software and client software, the former is in charge of image acquisition, encoding, transmission and storage, while the later is in charge of image processing, analysis and display. The lower software collects and encodes1080P HD images rapidly, and uploads the data to client for analysis through the network in real time, and storages the data to the external mass memory. At the same time the client software uses background subtraction, filtering blank images, tophat transformation and binaryzation to filter out images with the target microorganisms, divides the target areas, then identifies the target microorganisms with support vector machine, and updates statistics and marks, displays and stores the result. Experimental results indicate that the system software can carry out intelligent microbial identification in real time-and high accuracy; it can meet the needs of practical application.