Sea cucumber autonomous identification and grabbing method based on deep learning and binocular positioning

Abstract

The invention discloses a sea cucumber autonomous identification and grabbing method based on deep learning and binocular positioning. The sea cucumber autonomous identification and grabbing method comprises the following steps: performing underwater sea cucumber identification and positioning based on deep learning; acquiring sea cucumber spatial positioning information by utilizing binocular stereoscopic vision; and performing sea cucumber grabbing by using a PID control method. According to the invention, the GAN model is used to learn the characteristics of underwater sea cucumbers, and the generation network is used to generate sea cucumber samples, thereby effectively solving the problem of sea cucumber training sample insufficiency. According to the invention, mean filtering, medianfiltering and Wiener filtering are combined into a design filtering operator, so that the influence of non-uniform light, high turbidity, low visibility and the like on the image is solved. Accordingto the method, the convolutional neural network is utilized to learn and conclude the existing data, the sea cucumbers are accurately and quickly detected and two-dimensionally positioned, and a powerful guarantee is provided for subsequent spatial three-dimensional positioning and grabbing of the sea cucumbers. High-precision camera internal and external parameters are obtained, and accurate grabbing of the manipulator is guaranteed.

Description

Technical field [0001] The invention belongs to the field of intelligent underwater robots, and particularly relates to a sea cucumber autonomous identification and grasping method based on deep learning and binocular positioning. Background technique [0002] Sea cucumbers have extremely high nutritional value and medicinal effects. At this stage, sea cucumber fishing mainly relies on manual diving or remote manual underwater robots to obtain it. There are fatal defects such as low efficiency, limited underwater operation time, and high risk coefficient. [0003] With the development and popularization of artificial intelligence technology, intelligent underwater robots have been rapidly developed and applied in the field of marine autonomous fishing. Among them, the autonomous grasping of underwater sea cucumbers, sea urchins, scallops, abalones and other products is an important application field of intelligent underwater robots. Its core technologies mainly include rapid and ac...

AI Technical Summary

Problems solved by technology

[0005] In terms of rapid and accurate identification of sea cucumbers, the complexity and diversity of sample construction required for the deep learning training process, the impact of floating impurities and noise on the seabed, and low visibility on detection and identification have not been considered. Identify problems such as low detection accuracy, poor robustness, and low real-time performance;

[0006] In terms of three-dimensional positioning and matching of sea cucumbers, the influence of underwater refraction and multiple impurities on the calculation of the spatial position of sea cucumbers is not considered, resulting in large coordinate calculation errors in the world coordinate system of sea cucumbers;

[0007] At present, almost all the manipulators used for sea cucumber grasping are open-loop control. The manipulator only has two states of open and closed, and does not involve the information feedback of whether the sea cucumber is successfully grasped. Frequent

Images