A5090935711- Vehicle License Plate Recognition
- Optical measurement and interference techniques
- Industrial Vision Systems and Defect Detection
- Advanced Neural Network Applications
- Hand Gesture Recognition Systems
- Handwritten Text Recognition Techniques
- 3D Surveying and Cultural Heritage
- Video Surveillance and Tracking Methods
Hainan Medical University
2023
Shenyang Jianzhu University
2023
This paper proposes a multi-modality-based machine vision gap detection method, aiming at the problems of insufficient feature extraction, low accuracy point cloud segmentation, and poor edge fitting effect in traditional carbon fibre composite material methods. First, an improved sub-pixel method is proposed to extract more abundant features. Then, adaptive unified orientation designed achieve accurate segmentation centreline by enhancing curvature feature. Finally, innovative joint...
Aiming at the problems of low accuracy and slow recognition efficiency traditional traffic sign algorithm in complex environment, a deep learning method based on YOLOv5 is proposed. Firstly, Chinese data set TT100K randomly divided into training test set. Convolutional neural network YOLOv4 convolutional are used to train respectively set, so as build prediction model signs. Then trained validated Through evaluation experimental, it found that compared with model, has higher faster speed.
The license plate angle is unfixed, the vehicle position ununiform, and picture not sufficiently illuminated which leads to decrease of recognition accuracy. In order improve accuracy recognition, a deep learning-based method proposed. For location problem, YOLOv3 algorithm used. more capable recognizing small targets suitable for that requires precise positioning. problem character CNN plus multitask proposed recognition. experimental results show in this paper reaches 96%, intelligent realized.