Target Count Identification Using CNN Network Under Multiple Target

Radar detects the position and movement of objects by transmitting radio waves. It is often used to detect targets such as air defense interceptors, warships, and fighter jets, owing to its long maximum detection distance compared with optical devices such as cameras and infrared. Radar can measure the distance to an object and the speed and direction of movement of an object. However, owing to the limitations of its resolution, it is difficult to distinguish targets in situations in which multiple targets, such as missiles flying in clusters, approach simultaneously. In particular, when narrowband waveforms are used, some targets may not be detected when clustered targets are detected, making it difficult to distinguish between individual targets. To solve this problem, this study proposes a method to detect the number of targets in a clustered target group using a convolutional neural network to distinguish targets even in pulse waveforms with a narrow bandwidth, and verifies the performance of the method.