The invention discloses an underwater sensing method based on an artificial side line visual image. A side line system to be seen from a visual angle is provided, namely, classification and identification are carried out by importing a generated pressure difference image into a neural network so as to quickly and accurately carry out flow field perception and obstacle recognition, and a good foundation is provided for engineering application of an artificial side line system by improving the application limitation of an existing side line technology. The method comprises the following steps: 1) measuring pressure sensor array data under different working conditions; 2) arranging the pressure data measured in the step (1) into a pressure difference matrix; 3) drawing the pressure differencematrix obtained in the step (2) into a pressure difference RGB image, and 4) inputting the pressure difference RGB image obtained in the step (3) into a convolutional neural network for classification and identification to obtain each parameter of the underwater flow field and classify the parameters.