研究目的
To solve the problem about the inaccuracy and incompleteness of feature extraction and recognition in face recognition.
研究成果
The proposed method of multi-view ensemble learning in face recognition, combining multiple feature extraction and classification techniques, shows impressive recognition accuracy on face databases. Future work includes implementing the parallelism of the algorithm to compensate the complexity and further reduction in running time.
研究不足
The complexity of the algorithm and the need for further reduction in running time.
1：Experimental Design and Method Selection:
The method combines wavelet transform and edge detection for feature extraction, and KNN, WNN, and SVM for classification.
2：Sample Selection and Data Sources:
320 images from the ORL face database and some images from the FERET database were used.
3：List of Experimental Equipment and Materials:
Not explicitly mentioned.
4：Experimental Procedures and Operational Workflow:
Feature extraction using wavelet transform and edge detection, followed by classification using KNN, WNN, and SVM, and final integration using a voting strategy.
5：Data Analysis Methods:
Recognition rates were compared between the ensemble learning method and single classifiers.
独家科研数据包，助您复现前沿成果，加速创新突破
获取完整内容
