Improve Recognition Rate Research Articles

Synaptic plasticity and non-volatile memory characteristics in TiN-nanocrystal-embedded 3D vertical memristor-based synapses for neuromorphic systems.

Although vertical configurations for high-density storage require challenging process steps, such as etching high aspect ratios and atomic layer deposition (ALD), they are more affordable with a relatively simple lithography process and have been employed in many studies. Herein, the potential of memristors with CMOS-compatible 3D vertical stacked structures of Pt/Ti/HfOx/TiN-NCs/HfOx/TiN is examined for use in neuromorphic systems. The electrical characteristics (including I-V properties, retention, and endurance) were investigated for both planar single cells and vertical resistive random-access memory (VRRAM) cells at each layer, demonstrating their outstanding non-volatile memory capabilities. In addition, various synaptic functions (including potentiation and depression) under different pulse schemes, excitatory postsynaptic current (EPSC), and spike-timing-dependent plasticity (STDP) were investigated. In pattern recognition simulations, an improved recognition rate was achieved by the linearly changing conductance, which was enhanced by the incremental pulse scheme. The achieved results demonstrated the feasibility of employing VRRAM with TiN nanocrystals in neuromorphic systems that resemble the human brain.

Anomalous Behavior Detection Based on the Isolation Forest Model with Multiple Perspective Business Processes

Anomalous behavior detection in business processes inspects abnormal situations, such as errors and missing values in system execution records, to facilitate safe system operation. Since anomaly information hinders the insightful investigation of event logs, many approaches have contributed to anomaly detection in either the business process domain or the data mining domain. However, most of them ignore the impact brought by the interaction between activities and their related attributes. Based on this, a method is constructed to integrate the consistency degree of multi-perspective log features and use it in an isolation forest model for anomaly detection. First, a reference model is captured from the event logs using process discovery. After that, the similarity between behaviors is analyzed based on the neighborhood distance between the logs and the reference model, and the data flow similarity is measured based on the matching relationship of the process activity attributes. Then, the integration consistency measure is constructed. Based on this, the composite log feature vectors are produced by combining the activity sequences and attribute sequences in the event logs and are fed to the isolation forest model for training. Subsequently, anomaly scores are calculated and anomalous behavior is determined based on different threshold-setting strategies. Finally, the proposed algorithm is implemented using the Scikit-learn framework and evaluated in real logs regarding anomalous behavior recognition rate and model quality improvement. The experimental results show that the algorithm can detect abnormal behaviors in event logs and improve the model quality.

Dynamic Temperature Scaling in Contrastive Self-Supervised Learning for Sensor-Based Human Activity Recognition

The use of deep neural networks in sensor-based Human Activity Recognition has led to considerably improved recognition rates in comparison to more traditional techniques. Nonetheless, these improvements usually rely on collecting and annotating massive amounts of sensor data, a time-consuming and expensive task. In this paper, inspired by the impressive performance of Contrastive Learning approaches in Self-Supervised Learning settings, we introduce a novel method based on the SimCLR framework and a Transformer-like model. The proposed algorithm addresses the problem of negative pairs in SimCLR by using dynamic temperature scaling within a contrastive loss function. While the original SimCLR framework scales similarities between features of the augmented views by a constant temperature parameter, our method dynamically computes temperature values for scaling. Dynamic temperature is based on instance-level similarity values extracted by an additional model pre-trained on initial instances beforehand. The proposed approach demonstrates state-of-the-art performance on three widely used datasets in sensor-based HAR, namely MobiAct, UCI-HAR and USC-HAD. Moreover, it is more robust than the identical supervised models and models trained with constant temperature in semi-supervised and transfer learning scenarios.

Off-line English character recognition system

Purpose of the study: This paper aims to recognition of handwritten English characters in offline mode. It develops an efficient character recognition model avoiding large variations in handwriting by using better feature extraction techniques. Methodology: The samples of characters are preprocessed by applying a sequence of operations in succession like Thickening, Thresholding, Filtering, and Thinning. Efficient features like Gradient features and Zonal features have been extracted. Gradient features are helpful to find out stroke information in the character whereas Zonal features detail out local information in a more précised way. Hidden Markov Model is the classifier. Main Findings: Classification has been started with only a 5-state HMM model but it is observed that as the number of states of HMM model is increased, the corresponding recognition rate is also improved. Finally, with the 36 states HMM model we have got the expected result. This produces an overall average recognition rate of 92.6%. For the letters ‘A’ and ‘W’, the recognition rate is found to be very low, because of a lot of variations in writing style of these letters. Applications of this study: HMM is a flexible tool which is capable of absorbing variations in character images. The future works will be concentrated on improvement of recognition rate of such letters by finding some demarcating features and post processing. The proposed method can be well used in Natural Language Processing, Signature verification, Face recognition like other Pattern Recognition applications. Novelty/Originality of this study: Preprocessing uses Median filter which removes all stray marks in samples and hence avoids any possibility of false pixels. The combination of Gradient features and Zonal features leads to a recognition accuracy of 92.6% which may be used by researchers in any other domains for the purpose of classification. The application of HMM will motivate the readers to use it for better results of classification.

Facial Expression Recognition from a Single Face Image Based on Deep Learning and Broad Learning

With advances in computer vision and artificial intelligence technology, facial expression recognition research has become a prominent topic. Current research is grappling with how to enable computers to fully understand expression features and improve recognition rates. Most single face image datasets are based on the psychological classification of the six basic human expressions used for network training. By outlining the problem of facial recognition by comparing traditional methods, deep learning, and broad learning techniques, this review highlights the remaining challenges and future directions of deep learning and broad learning research. The deep learning method has made it easier and more effective to extract expression features and improve facial expression recognition accuracy by end-to-end feature learning, but there are still many difficulties in robustness and real-time performance. The broad learning system (BLS) is a broad network structure that is expanded by increasing the number of feature nodes and enhancement nodes appropriately to reinforce the structure and is also effective in facial expression recognition. However, outliers and noises in unbalanced datasets need BLS to solve in the future. Finally, we present several problems that still need to be addressed in facial expression recognition.

A Novel Piano Arrangement Timbre Intelligent Recognition System Using Multilabel Classification Technology and KNN Algorithm.

In this paper, melody and harmony are regarded as the task of machine learning, and a piano arranger timbre recognition system based on AI (Artificial Intelligence) is constructed by training a series of samples. The short-time Fourier transform spectrum analysis method is used to extract the piano timbre characteristic matrix, and the electronic synthesis of timbre recognition is improved by extracting the envelope function. Using the traditional multilabel classification method and KNN (K-nearest neighbor) algorithm, a combined algorithm of these two algorithms is proposed. The experimental results show that the detection rate increases from 61.3% to 70.2% after using the combined classification algorithm. The correct rate also increased from 40.3% to 48.9%, and the detection rate increased to 74.6% when the K value was set to 6. The experimental results show that, compared with the traditional classification algorithm, this algorithm has a certain improvement in recognition rate. Using this system to recognize the timbre of piano arrangement has a high recognition accuracy, which is worthy of further popularization and application.

Deep Learning-Based Segmentation of Peach Diseases Using Convolutional Neural Network.

Peach diseases seriously affect peach yield and people’s health. The precise identification of peach diseases and the segmentation of the diseased areas can provide the basis for disease control and treatment. However, the complex background and imbalanced samples bring certain challenges to the segmentation and recognition of lesion area, and the hard samples and imbalance samples can lead to a decline in classification of foreground class and background class. In this paper we applied deep network models (Mask R-CNN and Mask Scoring R-CNN) for segmentation and recognition of peach diseases. Mask R-CNN and Mask Scoring R-CNN are classic instance segmentation models. Using instance segmentation model can obtain the disease names, disease location and disease segmentation, and the foreground area is the basic feature for next segmentation. Focal Loss can solve the problems caused by difficult samples and imbalance samples, and was used for this dataset to improve segmentation accuracy. Experimental results show that Mask Scoring R-CNN with Focal Loss function can improve recognition rate and segmentation accuracy comparing to Mask Scoring R-CNN with CE loss or comparing to Mask R-CNN. When ResNet50 is used as the backbone network based on Mask R-CNN, the segmentation accuracy of segm_mAP_50 increased from 0.236 to 0.254. When ResNetx101 is used as the backbone network, the segmentation accuracy of segm_mAP_50 increased from 0.452 to 0.463. In summary, this paper used Focal Loss on Mask R-CNN and Mask Scoring R-CNN to generate better mAP of segmentation and output more detailed information about peach diseases.

An Adaptive Focal Loss Function Based on Transfer Learning for Few-Shot Radar Signal Intra-Pulse Modulation Classification

To solve the difficulty associated with radar signal classification in the case of few-shot signals, we propose an adaptive focus loss algorithm based on transfer learning. Firstly, we trained a one-dimensional convolutional neural network (CNN) with radar signals of three intra-pulse modulation types in the source domain, which were effortlessly obtained and had sufficient samples. Then, we transferred the knowledge obtained by the convolutional layer to nine types of few-shot complex intra-pulse modulation classification tasks in the target domain. We propose an adaptive focal loss function based on the focal loss function, which can estimate the parameters based on the ratio of hard samples to easy samples in the data set. Compared with other existing algorithms, our proposed algorithm makes good use of transfer learning to transfer the acquired prior knowledge to new domains, allowing the CNN model to converge quickly and achieve good recognition performance in case of insufficient samples. The improvement based on the focal loss function allows the model to focus on the hard samples while estimating the focusing parameter adaptively instead of tediously repeating experiments. The experimental results show that the proposed algorithm had the best recognition rate at different sample sizes with an average recognition rate improvement of 4.8%, and the average recognition rate was better than 90% for different signal-to-noise ratios (SNRs). In addition, upon comparing the training processes of different models, the proposed method could converge with the least number of generations and the shortest time under the same experimental conditions.

Random Graph-Based M-QAM Classification for MIMO Systems

Automatic modulation classification (AMC) has been identified to perform a key role to realize technologies such as cognitive radio, dynamic spectrum management, and interference identification that are arguably pivotal to practical SG communication networks. Random graphs (RGs) have been used to better understand graph behavior and to tackle combinatorial challenges in general. In this research article, a novel modulation classifier is presented to recognize M-Quadrature Amplitude Modulation (QAM) signals using random graph theory. The proposed method demonstrates improved recognition rates for multiple-input multiple-output (MIMO) and single-input single-output (SISO) systems. The proposed method has the advantage of not requiring channel/signal to noise ratio estimate or timing/frequency offset correction. Undirected RGs are constructed based on features, which are extracted by taking sparse Fourier transform (SFT) of the received signal. This method is based on the graph representation of the SFT of the 2nd, 4th, and 8th power of the received signal. The simulation results are also compared to existing state-of-the-art methodologies, revealing that the suggested methodology is superior.

FPN-Based Small Orange Fruit Detection From Farm Images With Occlusion

Fruit detection using deep learning is yielding very good performance, the goal of this work is to detect small fruits in images under these occlusion and overlapping conditions. The overlap among fruits and their occlusion can lead to false and missing detection, which decreases the accuracy and generalization ability of the model. Therefore, a small orange fruit recognition method based on improved Feature Pyramid Network was developed. To begin with, multi-scale feature fusion was used to fuse the detailed bottom features and high-level semantic features to detect small-sized orange to improve recognition rate. And then repulsion loss was used to take place of the original smooth L1 loss function. Besides, Soft non-maximum suppression was adopted to replace non-maximum suppression to screen the bounding boxes of orange to construct a recognition model of orange fruits. Finally, the network was trained and verified on the collected image data set. The results showed that compared with the traditional detection models, the mean average precision was improved from 79.7 to 82.8%.

A systematic literature review of speech emotion recognition approaches

Nowadays emotion recognition from speech (SER) is a demanding research area for researchers because of its wide real-life applications. There are many challenges for SER systems such as the availability of suitable emotional databases, identification of the relevant feature vector, and suitable classifiers. This paper critically analysed the literature on SER in terms of speech databases, speech features, traditional machine learning (ML) classifiers and DL approaches along with the areas for future directions. In recent years, there is a growing interest of researchers to use deep learning (DL) approaches for SER and get improvement in recognition rate. The focus of this review is on DL approaches for SER. A total of 152 papers have been reviewed from years 2000–2021. We have identified frequently used speech databases and related accuracies achieved using DL approaches. The motivations and limitations of DL approaches for SER are also summarized.

Target Recognition and Grabbing Positioning Method Based on Convolutional Neural Network

With the continuous reform of intelligent manufacturing, industrial production has gradually developed from automation to intelligence. The fusion of vision technology and industrial machines has become a hot research direction in current intelligent transformation. However, machines are not as flexible as humans when grabbing, and still have great limitations. Affected by various characteristics of target objects, such as shape, material, weight and other factors, as well as complex and changeable environmental factors, the research of machine grabbing still faces severe challenges. For the actual complex working conditions, the poor target detection effect leads to the inability to complete accurate grabbing, which affects the production efficiency. This paper proposes a grabbing system with convolutional neural network, which can achieve target detection, classification, positioning and grabbing tasks. First, by comparing the current mainstream target recognition and detection algorithms, select SSD that have both real-time performance and accuracy. Then make specific network structure improvements according to the detection requirements, and insert the Inception structure. At the same time optimize its loss function and nonmaximum suppression. The improved recognition rate is higher, and the target detection frame is closer to the real part, which greatly reduces the recognition error. Second, this research proposes an algorithm model for regional posture detection and grabbing positioning, which uses the output of the previous stage as input to perform posture detection and grabbing positioning of the grabbed target. In the network, the posture angle of the grabbing target is output in a classified manner, and the position coordinates of the grabbing point are output using a regression method. Experiments have proved that our method can perform efficient target recognition and grabbing positioning.

Machine learning-based automatic detection of novel coronavirus (COVID-19) disease.

The pandemic was announced by the world health organization coronavirus (COVID-19) universal health dilemma. Any scientific appliance which contributes expeditious detection of coronavirus with a huge recognition rate may be excessively fruitful to doctors. In this environment, innovative automation like deep learning, machine learning, image processing and medical image like chest radiography (CXR), computed tomography (CT) has been refined promising solution contrary to COVID-19. Currently, a reverse transcription-polymerase chain reaction (RT-PCR) test has been used to detect the coronavirus. Due to the moratorium period is high on results tested and huge false negative estimates, substitute solutions are desired. Thus, an automated machine learning-based algorithm is proposed for the detection of COVID-19 and the grading of nine different datasets. This research impacts the grant of image processing and machine learning to expeditious and definite coronavirus detection using CXR and CT medical imaging. This results in early detection, diagnosis, and cure for the accomplishment of COVID-19 as early as possible. Firstly, images are preprocessed by normalization to enhance the quality of the image and removing of noise. Secondly, segmentation of images is done by fuzzy c-means clustering. Then various features namely, statistical, textural, histogram of gradients, and discrete wavelet transform are extracted (92) and selected from the feature vector by principle component analysis. Lastly, k-NN, SRC, ANN, and SVM are used to make decisions for normal, pneumonia, COVID-19 positive patients. The performance of the system has been validated by the k (5) fold cross-validation technique. The proposed algorithm achieves 91.70% (k-Nearest Neighbor), 94.40% (Sparse Representation Classifier), 96.16% (Artificial Neural Network), and 99.14% (Support Vector Machine) for COVID detection. The proposed results show feature combination and selection improves the performance in 14.34 s with machine learning and image processing techniques. Among k-NN, SRC, ANN, and SVM classifiers, SVM shows more efficient results that are promising and comparable with the literature. The proposed approach results in an improved recognition rate as compared to the literature review. Therefore, the algorithm proposed shows immense potential to benefit the radiologist for their findings. Also, fruitful in prior virus diagnosis and discriminate pneumonia between COVID-19 and other pandemics.

The impact of collarette region-based convolutional neural network for iris recognition

Iris recognition is a biometric technique that reliably and quickly recognizes a person by their iris based on unique biological characteristics. Iris has an exceptional structure and it provides very rich feature spaces as freckles, stripes, coronas, zigzag collarette area, etc. It has many features where its growing interest in biometric recognition lies. This paper proposes an improved iris recognition method for person identification based on Convolutional Neural Networks (CNN) with an improved recognition rate based on a contribution on zigzag collarette area - the area surrounding the pupil - recognition. Our work is in the field of biometrics especially iris recognition; the iris recognition rate using the full circle of the zigzag collarette was compared with the detection rate using the lower semicircle of the zigzag collarette. The classification of the collarette is based on the Alex-Net model to learn this feature, the use of the couple (collarette/CNN) allows for noiseless and more targeted characterization and also an automatic extraction of the lower semicircle of the collarette region, finally, the SVM training model is used for classification using grayscale eye image data taken from (CASIA-iris-V4) database. The experimental results show that our contribution proves to be the best accurate, because the CNN can effectively extract the image features with higher classification accuracy and because our new method, which uses the lower semicircle of the collarette region, achieved the highest recognition accuracy compared with the old methods that use the full circle of collarette region.

An Enhanced Fuzzy Clustering to Pattern Recognition for Cloud Computing, by using Model Aggregation and Model Selection

Numerical schemes research on clustering models has been quite intensive in the past decade. Many models have been proposed to address the clustering tasks. Most clustering models are influenced by presentation order, complex shapes, architecture configuration, and learning instability. Hence, in the present study, a novel clustering-based method for cloud computing that provides an improvement in recognition rate, is described. The evaluation, based on 10-fold Cross-validation, showed that the proposed model, which is named BaggingCluster, yielded good results and performed better than Self Organizing Map and fuzzy Adaptive Resonance Theory. Experimental studies demonstrate that our model provides an efficient model for cloud computing.

Enhancing the Tracking of Seedling Growth Using RGB-Depth Fusion and Deep Learning.

The use of high-throughput phenotyping with imaging and machine learning to monitor seedling growth is a tough yet intriguing subject in plant research. This has been recently addressed with low-cost RGB imaging sensors and deep learning during day time. RGB-Depth imaging devices are also accessible at low-cost and this opens opportunities to extend the monitoring of seedling during days and nights. In this article, we investigate the added value to fuse RGB imaging with depth imaging for this task of seedling growth stage monitoring. We propose a deep learning architecture along with RGB-Depth fusion to categorize the three first stages of seedling growth. Results show an average performance improvement of correct recognition rate by comparison with the sole use of RGB images during the day. The best performances are obtained with the early fusion of RGB and Depth. Also, Depth is shown to enable the detection of growth stage in the absence of the light.

Face Recognition System Using Local Features Fusion for Multi-Masks

Face recognition is a relatively novel research field, and its application is closely related to numerous other areas. Moreover, it is emerging as a critical research theme due to its broad range of applications. Thus, many face recognition methods use a variety of feature extraction approaches. Nonetheless, the issue continues to be challenging, particularly identifying non-biological entities. This paper proposes an extended descriptor for local features of an effectual facial recognition system using a local directional pattern operator. This technique combines the Frei-Chen and Robinson masks’ strengths by fusion of the directional features of LDP for these two masks; this elicits a robust feature extraction method for distinguishing faces. Experimental results using the Yale database show that the extended descriptor considerably improved recognition rate and better performance than traditional local feature descriptors.

Robust children’s speech recognition in zero resource condition

User applications such as voice-based web search, online learning, and video gaming require an effective speech recognition module to take user commands. Nowadays, even children are frequently using such tools, especially for online learning and gaming. This has increased the demand for developing a noise-robust automatic speech recognition (ASR) system that can effectively transcribe children’s data under varied ambient conditions. However, automatic recognition of children’s speech is extremely challenging due to the insufficiency of data from child speakers in the majority of the languages across the world. Consequently, in this zero-resource condition, we are forced to decode children’s speech on systems trained using adults’ data. However, the acoustic mismatch between adults’ and children’s speech, such as differences in pitch, formant frequencies, and speaking-rates, leads to highly degraded recognition performance. To enhance the recognition rate under zero-resource conditions, we have explored the role of formant and duration-modification-based out-of-domain data augmentation in this paper. For that purpose, the formant frequencies of the adults’ speech data are upscaled using warping of linear predictive coding coefficients. On pooling original and formant modified adults’ speech data into training, the mismatch in formant locations is reduced leading to better recognition performance. Further improvement in recognition rate can be achieved by simultaneously modifying the duration as well as the formant frequencies of the training data. This case of out-of-domain data augmentation has also been studied in this work and found to yield added gains. In addition to data augmentation, a noise- and pitch-robust front-end acoustic feature extraction approach exploiting higher-order spectral analysis (simple and cross-bispectrum) is also proposed in this paper. The proposed features are noise-robust due to the inherent immunity of the bispectrum towards additive noises. An added advantage of bispectrum is reduced pitch sensitivity as demonstrated in this work. This, in turn, helps alleviate the aforementioned pitch-induced acoustic mismatch. The experimental evaluations presented in this paper demonstrate that the use of proposed acoustic features, as well as the out-of-domain data augmentation techniques, are highly suited for zero-resource children’s speech recognition tasks under clean and noisy conditions.

Recognition Rate Advancement and Data Error Improvement of Pathology Cutting with H-DenseUNet for Hepatocellular Carcinoma Image.

Due to the fact that previous studies have rarely investigated the recognition rate discrepancy and pathology data error when applied to different databases, the purpose of this study is to investigate the improvement of recognition rate via deep learning-based liver lesion segmentation with the incorporation of hospital data. The recognition model used in this study is H-DenseUNet, which is applied to the segmentation of the liver and lesions, and a mixture of 2D/3D Hybrid-DenseUNet is used to reduce the recognition time and system memory requirements. Differences in recognition results were determined by comparing the training files of the standard LiTS competition data set with the training set after mixing in an additional 30 patients. The average error value of 9.6% was obtained by comparing the data discrepancy between the actual pathology data and the pathology data after the analysis of the identified images imported from Kaohsiung Chang Gung Memorial Hospital. The average error rate of the recognition output after mixing the LiTS database with hospital data for training was 1%. In the recognition part, the Dice coefficient was 0.52 after training 50 epochs using the standard LiTS database, while the Dice coefficient was increased to 0.61 after adding 30 hospital data to the training. After importing 3D Slice and ITK-Snap software, a 3D image of the lesion and liver segmentation can be developed. It is hoped that this method could be used to stimulate more research in addition to the general public standard database in the future, as well as to study the applicability of hospital data and improve the generality of the database.

A deep convolutional neural network model for hand gesture recognition in 2D near-infrared images

The hand gesture recognition (HGR) process is one of the most vital components in human-computer interaction systems. Especially, these systems facilitate hearing-impaired people to communicate with society. This study aims to design a deep learning CNN model that can classify hand gestures effectively from the analysis of near-infrared and colored natural images. This paper proposes a new deep learning model based on CNN to recognize hand gestures improving recognition rate, training, and test time. The proposed approach includes data augmentation to boost training. Furthermore, five popular deep learning models are used for transfer learning, namely VGG16, VGG19, ResNet50, DenseNet121, and InceptionV3 and compared their results. These models are applied to recognize 10 different hand gestures for near-infrared images and 24 ASL hand gestures for colored natural images. The proposed CNN model, VGG16, VGG19, Resnet50, DenseNet121, and InceptionV3 models achieve recognition rates of 99.98%, 100%, 99.99%, 91.63%, 82.42% and 81.84%, respectively on near-infrared images. For colored natural ASL images, the models achieve recognition rates of 99.91%, 99.31%, 98.67%, 91.97%, 93.37%, and 93.21%, respectively. The proposed model achieves promising results spending the least amount of time.