Trainable Feature Extractor Research Articles

Deep learning based intelligent identification system for ripening stages of banana

Ripening of banana hands during handling, on board transit, shipping and storage leads to higher post-harvest loss and impede the trade. Identification of ripening is paramount importance to reduce loss. Bulk handlers and food processing industries requires automated non-destructive methods of ripening stage identification methodologies. This paper proposes a deep learning based non-destructive method of classification of banana fruit under four categories – unripe, under ripe, ripe and over ripe. A customized dataset was prepared with sufficient images in each class. A convolution neural network (CNN) combined with an eXtreme Gradient Boosting (XgBoost) algorithm (CNN-XgBoost) is introduced for the effective determination of the ripening stage of banana. CNN acts as the trainable feature extractor of the images and XgBoost acts as the identifier of ripening stage. Linear Discriminant Analysis (LDA) is incorporated in order to eliminate the need to have data augmentation or a huge data set. Thus, the proposed deep learning approach possesses capability to perform classification even with a smaller data set compared to conventional deep and machine learning techniques. The performance accuracy of the proposed duo is found to be 91.25 % and it is higher than that obtained with a Support Vector Classifier (SVC), Gaussian Naive Bayesian Classifier (GNB) or k-Nearest Neighbours (KNN) algorithms.

Classification of Mineral Foam Flotation Conditions Based on Multi-Modality Image Fusion

Accurate and rapid identification of mineral foam flotation states can increase mineral utilization and reduce the consumption of reagents. The traditional flotation process concentrates on extracting foam features from a single-modality foam image, and the accuracy is undesirable once problems such as insufficient image clarity or poor foam boundaries are encountered. In this work, a classification method based on multi-modality image fusion and CNN-PCA-SVM is proposed for work condition recognition of visible and infrared gray foam images. Specifically, the visible and infrared gray images are fused in the non-subsampled shearlet transform (NSST) domain using the parameter adaptive pulse coupled neural network (PAPCNN) method and the image quality detection method for high and low frequencies, respectively. The convolution neural network (CNN) is used as a trainable feature extractor to process the fused foam images, the principal component analysis (PCA) reduces feature data, and the support vector machine (SVM) is used as a recognizer to classify the foam flotation condition. After experiments, this model can fuse the foam images and recognize the flotation condition classification with high accuracy.

Hybrid CNN-SVR Blood Pressure Estimation Model Using ECG and PPG Signals.

Continuous blood pressure (BP) measurement is vital in monitoring patients' health with a high risk of cardiovascular disease. The complex and dynamic nature of the cardiovascular system can influence BP through many factors, such as cardiac output, blood vessel wall elasticity, circulated blood volume, peripheral resistance, respiration, and emotional behavior. Yet, traditional BP measurement methods in continuously estimating the BP are cumbersome and inefficient. This paper presents a novel hybrid model by integrating a convolutional neural network (CNN) as a trainable feature extractor and support vector regression (SVR) as a regression model. This model can automatically extract features from the electrocardiogram (ECG) and photoplethysmography (PPG) signals and continuously estimates the systolic blood pressure (SBP) and diastolic blood pressure (DBP). The CNN takes the correct topology of input data and establishes the relationship between ECG and PPG features and BP. A total of 120 patients with available ECG, PPG, SBP, and DBP data are selected from the MIMIC III database to evaluate the performance of the proposed model. This novel model achieves an overall Mean Absolute Error (MAE) of 1.23 ± 2.45 mmHg (MAE ± STD) for SBP and 3.08 ± 5.67 for DBP, all of which comply with the accuracy requirements of the AAMI SP10 standard.

Deep Residual Architecture Using Pixel and Feature Cues for View Synthesis and Temporal Interpolation

In this paper, we propose a deep residual architecture that can be used both for synthesizing high quality angular views in light fields and temporal frames in classical videos. The proposed framework consists of an optical flow estimator optimized for view synthesis, a trainable feature extractor and a residual convolutional network for pixel and feature-based view reconstruction. Among these modules, the fine-tuning of the optical flow estimator specifically for the view synthesis task yields scene depth or motion information that is well optimized for the targeted problem. In cooperation with the end-to-end trainable encoder, the synthesis block employs both pixel-based and feature-based synthesis with residual connection blocks, and the two synthesized views are fused with the help of a learned soft mask to obtain the final reconstructed view. Experimental results with various datasets show that our method performs favorably against other state-of-the-art (SOTA) methods with a large gain for light field view synthesis. Furthermore, with a little modification, our method can also be used for video frame interpolation, generating high quality frames compared with SOTA interpolation methods.

A Hybrid DCNN-SVM Model for Classifying Neonatal Sleep and Wake States Based on Facial Expressions in Video.

Sleep is a natural phenomenon controlled by the central nervous system. The sleep-wake pattern, which functions as an essential indicator of neurophysiological organization in the neonatal period, has profound meaning in the prediction of cognitive diseases and brain maturity. In recent years, unobtrusive sleep monitoring and automatic sleep staging have been intensively studied for adults, but much less for neonates. This work aims to investigate a novel video-based unobtrusive method for neonatal sleep-wake classification by analyzing the behavioral changes in the neonatal facial region. A hybrid model is proposed to monitor the sleep-wake patterns of human neonates. The model combines two algorithms: deep convolutional neural network (DCNN) and support vector machine (SVM), where DCNN works as a trainable feature extractor and SVM as a classifier. Data was collected from nineteen Chinese neonates at the Children's Hospital of Fudan University, Shanghai, China. The classification results are compared with the gold standard of video-electroencephalography scored by pediatric neurologists. Validations indicate that the proposed hybrid DCNN-SVM model achieved reliable performances in classifying neonatal sleep and wake states in RGB video frames (with the face region detected), with an accuracy of 93.8±2.2% and an F1-score 0.93±0.3.

Incremental Concept Learning via Online Generative Memory Recall.

The ability to learn more concepts from incrementally arriving data over time is essential for the development of a lifelong learning system. However, deep neural networks often suffer from forgetting previously learned concepts when continually learning new concepts, which is known as the catastrophic forgetting problem. The main reason for catastrophic forgetting is that past concept data are not available, and neural weights are changed during incrementally learning new concepts. In this article, we propose an incremental concept learning framework that includes two components, namely, ICLNet and RecallNet. ICLNet, which consists of a trainable feature extractor and a dynamic concept memory matrix, aims to learn new concepts incrementally. We propose a concept-contrastive loss to alleviate the magnitude of neural weight changes and mitigate the catastrophic forgetting problems. RecallNet aims to consolidate old concepts memory and recall pseudo samples, whereas ICLNet learns new concepts. We propose a balanced online memory recall strategy to reduce the information loss of old concept memory. We evaluate the proposed approach on the MNIST, Fashion-MNIST, and SVHN data sets and compare it with other pseudorehearsal-based approaches. Extensive experiments demonstrate the effectiveness of our approach.

MedGAN: Medical image translation using GANs

Image-to-image translation is considered a new frontier in the field of medical image analysis, with numerous potential applications. However, a large portion of recent approaches offers individualized solutions based on specialized task-specific architectures or require refinement through non-end-to-end training. In this paper, we propose a new framework, named MedGAN, for medical image-to-image translation which operates on the image level in an end-to-end manner. MedGAN builds upon recent advances in the field of generative adversarial networks (GANs) by merging the adversarial framework with a new combination of non-adversarial losses. We utilize a discriminator network as a trainable feature extractor which penalizes the discrepancy between the translated medical images and the desired modalities. Moreover, style-transfer losses are utilized to match the textures and fine-structures of the desired target images to the translated images. Additionally, we present a new generator architecture, titled CasNet, which enhances the sharpness of the translated medical outputs through progressive refinement via encoder-decoder pairs. Without any application-specific modifications, we apply MedGAN on three different tasks: PET-CT translation, correction of MR motion artefacts and PET image denoising. Perceptual analysis by radiologists and quantitative evaluations illustrate that the MedGAN outperforms other existing translation approaches.

Fault detection in flotation processes based on deep learning and support vector machine

Effective fault detection techniques can help flotation plant reduce reagents consumption, increase mineral recovery, and reduce labor intensity. Traditional, online fault detection methods during flotation processes have concentrated on extracting a specific froth feature for segmentation, like color, shape, size and texture, always leading to undesirable accuracy and efficiency since the same segmentation algorithm could not be applied to every case. In this work, a new integrated method based on convolution neural network (CNN) combined with transfer learning approach and support vector machine (SVM) is proposed to automatically recognize the flotation condition. To be more specific, CNN function as a trainable feature extractor to process the froth images and SVM is used as a recognizer to implement fault detection. As compared with the existed recognition methods, it turns out that the CNN-SVM model can automatically retrieve features from the raw froth images and perform fault detection with high accuracy. Hence, a CNN-SVM based, real-time flotation monitoring system is proposed for application in an antimony flotation plant in China.

Genetic algorithm‐optimised structure of convolutional neural network for face recognition applications

Proposing a proper method for face recognition is still a challenging subject in biometric and computer vision applications. Although some reliable systems were introduced under relatively controlled conditions, their recognition rate is not satisfactory in the general settings. This is especially true when there are variations in pose, illumination, and facial expression. To alleviate these problems, a hybrid face recognition system is proposed which benefits from the superiority of both convolutional neural network (CNN) and support vector machine (SVM). To this end, first a genetic algorithm is employed to find the optimum structure of CNN. Then, the performance of the system is improved by replacing the last layer of CNN with an ensemble of SVMs. Finally, using concepts of error correction, decision is made. The potential of CNN as a trainable feature extractor provides a flexible recognition system that can recognise faces with variations in pose and illumination. Simulation results show that the system achieves good recognition rate and is robust against variations in terms of facial expressions, occlusion, noise, and illuminations.

A novel hybrid CNN–SVM classifier for recognizing handwritten digits

This paper presents a hybrid model of integrating the synergy of two superior classifiers: Convolutional Neural Network (CNN) and Support Vector Machine (SVM), which have proven results in recognizing different types of patterns. In this model, CNN works as a trainable feature extractor and SVM performs as a recognizer. This hybrid model automatically extracts features from the raw images and generates the predictions. Experiments have been conducted on the well-known MNIST digit database. Comparisons with other studies on the same database indicate that this fusion has achieved better results: a recognition rate of 99.81% without rejection, and a recognition rate of 94.40% with 5.60% rejection. These performances have been analyzed with reference to those by human subjects.

A trainable feature extractor for handwritten digit recognition

This article focuses on the problems of feature extraction and the recognition of handwritten digits. A trainable feature extractor based on the LeNet5 convolutional neural network architecture is introduced to solve the first problem in a black box scheme without prior knowledge on the data. The classification task is performed by support vector machines to enhance the generalization ability of LeNet5. In order to increase the recognition rate, new training samples are generated by affine transformations and elastic distortions. Experiments are performed on the well-known MNIST database to validate the method and the results show that the system can outperform both SVMs and LeNet5 while providing performances comparable to the best performance on this database. Moreover, an analysis of the errors is conducted to discuss possible means of enhancement and their limitations.