摘要
水下图像目标检测与陆地环境目标检测相比,水下场景表现出更大的复杂性。这会导致捕获图像中的背景颜色发生明显变化,从而影响基于深度学习的通用目标检测模型的检测精度。为了应对这一挑战并提高水下图像目标检测的精度,设计一种CBAM-CRMS注意力机制模块,将该模块嵌入YOLOv7检测模型,通过实验验证了改进算法的优越性能。
The underwater image target detection is more complicated than the terrestrial environment target detection.This results in a significant change in the background color in the captured image,which affects the detection accuracy of the universal object detection model based on deep learning.In order to meet this challenge and improve the accuracy of underwater image target detection,a CBAMCRMS attentional mechanism module is designed in this paper,and the module is embedded in the YOLOv7 detection model.The superior performance of the improved algorithm is verified through experiments.
作者
邹敏
ZOU Min(School of Advanced Manufacturing,Fuzhou University,Quanzhou 362200,China)
出处
《电视技术》
2023年第12期55-59,共5页
Video Engineering