Subjects: Computer Science >> Integration Theory of Computer Science submitted time 2018-11-29 Cooperative journals: 《计算机应用研究》
Abstract： Aimed-at lung nodule Benign/Malignant classification, an effective grey scale density distribution feature extraction algorithm which was combined with pattern recognition models to evaluate the classification system was proposed. The proposed feature extraction algorithm first collected a large number of blocks from lung tumor images and determined the distance matrix by calculating the relationships among the image blocks. Then, K-means clustering methods was used to classify the current image blocks and obtained 10 cluster centers. After that, calculated the distribution density features by mapping CT value of nodule image pixels with the 10 cluster centers and extracted a 10-dimensional feature vector. Finally, the extracted feature vectors were divided into training and testing set to identify lung adenocarcinomas risk levels by Random Forest classification model. The classification framework was evaluated in LIDC-IDRI dataset, the average accuracy reached to 0.9008. The proposed method outperforms the most recent techniques, and the experimental results show great robustness of the proposed method for different lung CT image datasets.
Subjects: Computer Science >> Integration Theory of Computer Science submitted time 2018-08-13 Cooperative journals: 《计算机应用研究》
Abstract： Fatigue driving is one of the main causes of traffic accidents, particularly for drivers of large vehicles like buses and heavy trucks. For designing driver fatigue monitoring systems, the visual features based techniques is one of the most effective approaches. This paper proposed a hierarchical convolutional neural network model with multi-scale pooling for vision-based fatigue detection system. The first step is face detection and extraction of eye and mouse regions by deep learning model—MTCNN. In order to solve the problem of characterization and recognition of eye and mouth regions, this paper proposed a multi-scale pooling model（MSP）based on ResNet to train different states of eye and mouth. In real-time detection, the paper recognized the states of eye and mouth by the pre-trained convolutional neural network model. Finally, this paper detect fatigue through the PERCLOS and the frequency of open mouth (FOM) . The experimental results show that the proposed algorithm has high detection accuracy, real-time performance and high robustness to complex environments.
Hits
Hits
Downloads
Comment