Artificial neural network technology to identify ice slurry density of the Yellow River

摘要

This study using computer image processing and artificial neural network sensor technologies constructs a method of identifying ice slurry density based on the value of ice color image. The method is applied to the Jinan section of the Yellow River through the ice image acquisition, R/G color extraction, network learning and training, the final output target value of ice or water, and the actual image as an example identification checking. The collected 96 × 128 pixel images are input to the trained neural network model and the density of the image ice slurry is calculated at 66.18%. The results show that the method has a high computational speed, good agreement with the actual results of the feature, and realizes the purpose of automatically recognizing ice slurry density of the Yellow River on the computer platform.