In this paper we design a fast fabric defect detection framework (Fast-DDF) based on gray histogram back-projection, which adopts end to end multi-convoluted network model to realize defect classification. First, the ...
详细信息
In this paper we design a fast fabric defect detection framework (Fast-DDF) based on gray histogram back-projection, which adopts end to end multi-convoluted network model to realize defect classification. First, the back-projection image is established through the gray histogram on fabric image, and the closing operation and adaptive threshold segmentation method are performed to screen the impurity information and extract the defect regions. Then, the defect images segmented by the Fast-DDF are marked and normalized into the multi-layer convolutional neural network for training. Finally, in order to solve the problem of difficult adjustment of network model parameters and long training time, some strategies such as batch normalization of samples and network fine tuning are proposed. The experimental results on the TILDA database show that our method can deal with various defect types of textile fabrics. The average detection accuracy with a higher rate of 96.12% in the database of five different defects, and the single image detection speed only needs 0.72s.
As a promising imaging technique, electrical impedance tomography (EIT) is used to reflect the conductivity distribution variation of human tissues. Different from the lung EIT, the application of the craniocerebral E...
详细信息
As a promising imaging technique, electrical impedance tomography (EIT) is used to reflect the conductivity distribution variation of human tissues. Different from the lung EIT, the application of the craniocerebral EIT is challenging. This is attributed to the fact that the skull with high resistivity greatly restricts the injected current from flowing into the brain tissue. Consequently, the measured voltage signal is very weak causing poor reconstructed images. To solve this problem, a new strategy based on a multi-layer convolutional neural network (CNN) is proposed for weak voltage signal enhancement. Voltage measurements from the three-layer head model and the single-layer head model perform as the input and the output of the network respectively. The trained network is supposed to enhance the voltage data of the three-layer head model. To test the performance of the proposed method, voltage data processed by the multi-layer CNN is compared with the single-layer voltage data. Besides, comparisons are also made in the case of noise interruption and when the skull thickness varies. The results demonstrate that the processed voltage data is almost consistent with the single-layer voltage data. Compared with the image reconstruction with the three-layer voltage data, there is a large improvement when using the proposed method.
The decision-making process in healthcare monitoring systems has extensively used deep learning and the Internet of Things (IoT). Among the new uses in today's procedures is disease prediction. A difficult challen...
详细信息
The decision-making process in healthcare monitoring systems has extensively used deep learning and the Internet of Things (IoT). Among the new uses in today's procedures is disease prediction. A difficult challenge in computer-aided diagnosis (CAD) is lung cancer prediction, which is addressed in this paper's solution using deep learning and IoT. IoT medical devices send disease-related data to the server, as lung cancer is a hazardous medical condition that must be detected faster. Following processing, a multi-layer convolutional neural network (ML-CNN) model is used to classify the medical data into benign and malignant groups. Enhanced Particle swarm optimization (EPSO) is also used to enhance learning capacity (accuracy and loss). Medical data from the Internet of Medical Things (IoMT), including sensor data and Computed Tomography (CT) scan results, is used in this stage. The information from sensors and IoMT devices' picture data is collected for this purpose, and classification operations are then performed. The suggested method's accuracy, precision, sensitivity, specificity, F-score, and computation time are compared to well-known current approaches such as the Support Vector Machine (SVM), Probabilistic neuralnetwork (PNN), and convolutionalneuralnetwork (CNN). Linear Imaging and Self-Scanning Sensor (LISS) and Lung Image Database Consortium (LIDC) datasets were the two lung datasets used for this performance evaluation. Trial results indicate that the recommended strategy may aid in the timely and accurate identification of lung cancer in radiologists compared to other techniques. The efficacy of the suggested ML-CNN was examined through Python analysis. The results showed that the accuracy was superior to the number of instances, the precision was superior to the number of cases, and the sensitivity was superior to several instances, the F-score was superior to the number of cases, the error rate was inferior to the number of cases, and the computat
Sign Language Recognition (SLR) helps to bridge the gap between ordinary and hearing-impaired people. But various difficulties and challenges are faced by SLR system during real-time implementation. The major complexi...
详细信息
Sign Language Recognition (SLR) helps to bridge the gap between ordinary and hearing-impaired people. But various difficulties and challenges are faced by SLR system during real-time implementation. The major complexity associated with SLR is the inability to provide a consistent recognition process and it shows lesser recognition accuracy. To handle this issue, this research concentrates on adopting the finest classification approach to provide a feasible end-to-end system using deep learning approaches. This process transforms sign language into the voice for assisting the people to hear the sign language. The input is taken from the ROBITA Indian Sign Language Gesture Database and some essential pre-processing steps are done to avoid unnecessary artefacts. The proposed model is incorporated with the encoder multi-layer convolutional neural networks (ML-CNN) for evaluating the scalability, accuracy of the end-to-end SLR. The encoder analyses the linear and non-linear features (higher level and lower level) to improve the quality of recognition. The simulation is carried out in a MATLAB environment where the performance of the ML-CNN model outperforms the existing approaches and establishes the trade-off. Some performance metrics like accuracy, precision, F-measure, recall, Matthews Correlation Coefficient (MCC), Mean Absolute Error (MAE) are evaluated to show the significance of the model. The prediction accuracy of the proposed ML-CNN with encoder is 87.5% in the ROBITA sign gesture dataset and it's increased by 1% and 3.5% over the BLSTM and HMM respectively.
multi-label text classification is a widely studied subtask in the field of natural language processing. Unlike single-label text classification, multi-label text classification involves assigning multiple labels to a...
详细信息
Protein interactions play an essential role in studying living systems and life phenomena. A considerable amount of literature has been published on analyzing and predicting protein interactions, such as support vecto...
详细信息
Protein interactions play an essential role in studying living systems and life phenomena. A considerable amount of literature has been published on analyzing and predicting protein interactions, such as support vector machine method, homology-based method and similarity-based method, each has its pros and cons. Most existing methods for predicting protein interactions require prior domain knowledge, making it difficult to effectively extract protein features. Single method is dissatisfactory in predicting protein interactions, declaring the need for a comprehensive method that combines the advantages of various methods. On this basis, a deep ensemble learning method called EnAmDNN (Ensemble Deep neuralnetworks with Attention Mechanism) is proposed to predict protein interactions which is an appropriate candidate for comprehensive learning, combining multiple models, and considering the advantages of various methods. Particularly, it encode protein sequences by the local descriptor, auto covariance, conjoint triad, pseudo amino acid composition and combine the vector representation of each protein in the protein interaction network. Then it takes advantage of the multi-layer convolutional neural networks to automatically extract protein features and construct an attention mechanism to analyze deep-seated relationships between proteins. We set up four different structures of deep learning models. In the ensemble learning model, second layer data sets are generated with five-fold cross validation from basic learners, then predict the protein interaction network by combining 16 models. Results on five independent PPI data sets demonstrate that EnAmDNN achieves superior prediction performance than other comparing methods.
暂无评论