Abstract:

A range image of imaging laser radar is usually disturbed by dropout and range anomalies. The data collection through real laser radar is time-consuming and is difficult to control the noise level. To solve these problems, a new simulation method for range images with noise is proposed. According to the noise formation, in laser radar range image, an accurate laser radar system model is constructed, which includes major affecting factors such as speckle, atmospheric turbulence and receiver noise. The range images with noise under various scenes are obtained by using MATLAB software. Simulation results reveal that speckle, atmospheric turbulence and receiver noise are major factors to cause the dropout and range anomalies when using the proper detection strategy; The receiver noise alone has a limited effect on range anomalies; The dropout and range anomalies both exist in background and target region.