The present invention relates to the technical field of submerged oil detection, and specifically relates to a deep convolutional neural network-based submerged oil sonar detection image recognition method. A sonar detection image preprocessing module and a deep convolutional neural network-based submerged oil target recognition module are comprised. The present invention introduces a deep convolutional neural network algorithm to the detection and recognition of a submerged oil, solves the key problems of automatic recognition of a submerged oil target, automatic positioning of a leakage position, automatic estimation of the pollution area of the submerged oil and the like, provides a basis for emergency decision and disposal of marine oil spill accidents, improves the level of a submerged oil detection and recognition technology in China, provides technical support for national marine safety and petroleum safety, and has important engineering significance and an application value.
展开▼