A method and system for quantitative detection of near-seafloor benthic organisms based on deep learning

By adopting the YOLOv8 network model in offshore benthic bioquantitative detection, combined with SPD-Conv and BiFormer modules, the problems of low detection accuracy and speed in the existing technology are solved, and higher detection accuracy and faster detection speed are achieved .

CN116664545BActive Publication Date: 2024-12-27YANCHENG INST OF TECH +1
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202310735663.7
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2023-06-20
Publication Date
2024-12-27
Estimated Expiration
2043-06-20

AI Technical Summary

Technical Problem

The accuracy of the detection results of existing deep learning technologies in offshore benthic bioquantitative detection is affected by the complexity of the seabed environment, and the convolutional neural network has poor adaptability to interference characteristics and has a low detection speed.

Method used

The YOLOv8 network model is adopted, combined with the SPD-Conv module and BiFormer module, and replace the backbone network convolution module and pooling layer, enhance the image feature processing and generalization capabilities of the model and improve the detection speed.

Benefits of technology

It improves the accuracy and detection speed of quantitative detection of offshore benthic organisms, and enhances the adaptability and generalization capabilities of the model.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN116664545B_ABST
    Figure CN116664545B_ABST
Patent Text Reader

Abstract

The present invention provides a method and system for quantitative detection of near-sea benthic organisms based on deep learning, wherein the system comprises: step 1: obtaining an optical image data set of near-sea benthic organisms; step 2: dividing the optical image data set to obtain a training set and a test set; step 3: setting a YOLOv8 network model; step 4: determining a near-sea benthic organism quantitative detection model based on deep learning technology, according to the YOLOv8 network model, the training set and the test set; step 5: obtaining an image to be detected; step 6: inputting the image to be detected into the near-sea benthic organism quantitative detection model to obtain a near-sea benthic organism quantitative detection result. The present invention provides a method and system for quantitative detection of near-sea benthic organisms based on deep learning, wherein a YOLOv8 network model is set to train a near-sea benthic organism quantitative detection model, thereby improving the processing capability of the near-sea benthic organism quantitative detection model for image features, achieving a higher detection accuracy, and enhancing the generalization capability of the model. In addition, the detection speed is faster.
Need to check novelty before this filing date? Find Prior Art