2018 IEEE Conference on Data Science: Challenges of Digital Transformation (DSDT)

5 / 2018-03-28 22:10:43

A JOINT APPLICATION OF FUZZY LOGIC APPROXIMATION AND A DEEP LEARNING NEURAL NETWORK TO BUILD FISH CONCENTRATION MAPS BASED ON SONAR DATA

Fuzzy Logic,Neural network

In this paper we propose a novel method for obtaining topographic maps of lakes, maps of fish concentration and a map of predator location based on the results of intelligent sonar data processing. The method uses an effective algorithm for the detection of fishes and other objects on sonar images based on the following steps: input frame separating into overlapping blocks, blocks-processing using convolutional neural networks (CNN) YOLO v2, and merging extracted bounding boxes around one object. To construct maps of the distribution of features along the lake, we propose a new method for constructing the approximation of GPS-referenced CNN results based on the original implementation of fuzzy logic.