基于yolo-Idstd算法的红外弱小目标检测

Monocular camera-based large field-of-view depth estimation for mobile machining of large aerospace components

“…This model adds an attention mechanism to the traditional Yolov8 neural network and incorporates CNN convolutional neural network for deep regression analysis [5] . The proposed Yolov8sim-CNN neural network can realize the operation of classification, contour segmentation and image binarization for images taken by monocular camera in real time, and then predict the obtained information for feature extraction and finally realize the prediction of workpiece depth, and the structure of this neural network is shown in Figure 3.…”

A lightweight target detection network: ghost-YOLONet

“…The literature [10] adds Max Module to the lightweight network YOLO v4-tiny to capture more of the target's main features and combines it with a multi-scale fusion module to meet the requirements of real-time detection scenarios. The literature [11] adds Focus structure and multi-stage convolutional pooling module to the frontend of the feature extraction network of the YOLO model, proposing a lightweight algorithm for infrared weak target detection that reduces the reliance on hardware platforms.…”

Study on performance evaluation of intelligent photovoltaic defect detection by electroluminescence imaging

“…Traditionally, PV defect detection based on EL images has heavily relied on manual visual assessments. However, with the continual advancement [4][5][6][7][8][9][10][11] in image classification, localization, and segmentation tasks, machine learning and deep learning technologies have begun to find widespread application in the domain of PV defect detection.…”