Correct Recognition Rate Research Articles

VMD-based vibrating fiber system intrusion signal recognition

For the Sagnac fiber perimeter security system, the recognition rate of intrusion signal is not high. This paper proposes a new endpoint detection algorithm and recognition algorithm to effectively improve the recognition rate of intrusion signals. The short-term logarithmic energy and spectral entropy characteristics are combined to form a new endpoint detection algorithm to improve the accuracy of endpoint detection. The recognition algorithm uses variational mode decomposition to extract the spectral entropy, energy ratio and kurtosis of the eigenmode function. Including the multi-dimensional features of time domain and frequency domain, using the uncertainty to reduce the feature dimension. The support vector machine is selected to realize the intrusion signal recognition. The experimental results show that the proposed recognition algorithm can effectively identify the tapping, walking and stone throwing signals. The correct recognition rate reached 98.0 %.

Detection of Moldy Cores in Apples with Near-Infrared Transmission Spectroscopy Based on Wavelet and BP Network

Moldy cores in apples are not initially obvious from the outside of the fruit, so developing methods to detect moldy cores is an important area of research in the apple industry. The objective of this study was to improve the ability of near-infrared spectrometry to detect moldy cores in apples. Transmission spectra were recorded for 200 apple samples in the range of 200–1100[Formula: see text]nm, and 140 and 60 samples were randomly selected as training and test sets, respectively. Signal de-noising was performed by wavelet thresholding based on the results of orthogonal experiments. The best wavelengths for discriminating between healthy and diseased apples were selected by a successive projection algorithm (SPA). The extracted wavelengths were used as the input in a back propagation artificial neural network (BP-ANN). Through these experiments, this study compared the correct recognition rates using different ratios of training to test numbers in the model, and functions in the hidden and output layers of the BP-ANN. The proposed method achieved the highest accuracies of 95.00% and 95.71% for the test and training sets, respectively. This method could be used to develop a portable instrument for detecting moldy cores in apples.

Real-time Recognition Method of The Ship Object Based on Video Image

Target recognition has always been a hot topic in the field of computer vision. In the military field, computer vision has a good application in the research of ship target detection methods based on SAR images, but this type of research is limited to the detection of ship targets, and there is no further identification of ship types. Shipborne tracking radar parasitic TV can obtain clear images of sea surface targets in sea and air background. In order to improve the combat capability of ship combat systems, this paper proposes a video-based ship target recognition method, based on the identification of ship targets. This paper takes the hand-labeled TV video as the research object, and uses the Camshift and Kalman filter to improve the Faster R-CNN calculation framework for the automatic recognition and tracking of the video image ship target in the sea and sky background. The experimental results show that the correct recognition rate of the target model of the ship based on this method is above 90%.

Subclass Separation of White Blood Cell Images Using Convolutional Neural Network Models

The white blood cells produced in the bone marrow and lymphoid tissue known as leucocytes are an important part of the immune system to protect the body against foreign invaders and infectious disease. These cells, which do not have color, have a few days or several weeks of life. A lot of clinic experience is required for a doctor to detect the amount of white blood cells in human blood and classify it. Thus, early and accurate diagnosis can be made in the formation of various disease types, including infection on the immune system, such as anemia and leukemia, while evaluating and determining the disease of a patient. The white blood cells can be separated into four subclasses, such as Eosinophil, Lymphocyte, Monocyte, and Neutrophil. This study focuses on the separation of the white blood cell images by the classification process using convolutional neural network models, which is a deep learning model. A deep learning network, which is slow in the training step due to the complex architecture, but fast in the test step, is used for the feature extraction instead of intricate methods. For the subclass separation of white blood cells, the experimental results show that the AlexNet architecture gives the correct recognition rate among the convolutional neural network architectures tested in the study. Various classifiers are performed on the features derived from the AlexNet architecture to evaluate the classification performance. The best performance in the classification of white blood cells is given by the quadratic discriminant analysis classifier with the accuracy of 97.78 %.

Multifractal Spectrum Identification of Partial Discharge in GIS Based on Optical Detection

Partial discharge (PD) pattern recognition is an effective means to accurately obtain the correspondence between PD features and insulation defects. Optical detection for PD has strong anti-interference ability. Therefore, this paper carried out research on the pattern recognition of PD optical signal. In this paper, the PD optical signal under four typical defects of gas-insulated switchgears (GIS) are obtained by fluorescent fiber sensing system, and the four kinds of grayscale images are established. The multifractal spectrum features of the grayscale images of the optical signals are extracted. By analysing the physical meaning of each feature information in the multifractal spectrum, the feature components for PD pattern recognition are extracted, and the improved conjugate gradient back propagation (BP) neural network algorithm is selected as the classifier. This paper show that the multifractal spectrum can effectively describe the degree of unevenness of PD grayscale images and the geometric features of different level. The correct recognition rate of all kinds of defects by the multifractal spectrum is more than 87%, which is better than the identification of the box dimension and information dimension as the feature components. It lays the foundation for the future application of the optical detection.

Fatal obstructive asphyxia: Trans-pulmonary density gradient characteristic as relevant identifier in postmortem CT

Abstract To investigate effects of obstructive asphyxia on the postmortem computed tomography (PMCT) appearance of lungs, we matched 24 cases of obstructive asphyxia with age-sex matched controls. The initial qualitative impression was that obstructive asphyxia typically appeared to correlate with absence or reduced postmortem hypostasis as well as increased pulmonary radiotranslucency. Categories initially included within obstructive asphyxia were fatal aspiration of gastric contents into lungs, fatally blocked trachea or bronchi due to foreign body, fatal ligature strangulation, fatal manual strangulation and smothering. Method: We manually placed density profile vectors, with start at positionally dependent locations of each lung to derive PMCT densities for multivariate statistics. As fatal gastric content aspiration cases did not significantly differ from the controls, they were re-grouped into the control group for further testing. Results: Quantitatively, the obstructive asphyxia case group featured reduced or absent postmortem hypostasis in the lungs and an overall lower density. Linear multivariate statistics (general linear models, linear discriminant analysis) performed relatively poorly, whereas ”shallow” neural networks yielded significantly better discrimination. With disjunct training and test sets, neural networks achieved correct recognition rates (CRR) of obstructive asphyxia cases of over 95%. Conclusions: Within a more narrowly defined collective of obstructive asphyxia, pulmonary PMCT data appears to contain sufficient information to justify its use in a screening process. This questions the value of an exclusively death scene focused decision making for further investigations and may justify a wider use of PMCT. Conversely, obstructive asphyxia remains difficult to diagnose conclusively based on PMCT alone, so a full autopsy including toxicological analysis are recommended in such cases.

Static object imaging features recognition algorithm in dynamic scene mapping

In dynamic scene monocular visual SLAM, it is important to recognize static object imaging features for mapping. The optical flow of noise and the object moving in the same orientations would have the same orientations of static feature for two frames image, that would disturb static feature point recognition. Because the translational motion optical flow orientations method judges with optical flow of single feature point and translational motion orientations, it is not reliable for static feature point recognition. Therefore, a 3D motion segmentation combined for static object imaging features recognition is proposed, which clusters the same optical flow of feature points in 3D motion subspace, including orientations and amplitude of translation and rotation. The simulation results show that the proposed static object imaging features recognition method improves the recognition correct rate and reliability of static features. So, it is effective for static object imaging features recognition in dynamic scene mapping.

Unsteady State Lightweight Iris Certification Based on Multi-Algorithm Parallel Integration

Aimed at the one-to-one certification problem of unsteady state iris at different shooting times, a multi-algorithm parallel integration general model structure is proposed in this paper. The iris in the lightweight constrained state affected by defocusing, deflection, and illumination is taken as the research object, the existing algorithms are combined into the model structure effectively, and a one-to-one certification algorithm for lightweight constrained state unsteady iris was designed based on multi-algorithm integration and maximum trusted decision. In this algorithm, a sufficient number of iris internal feature points from the unstable state texture were extracted as effective iris information through the image processing layer composed of various filtering processing algorithms, thereby eliminating defocused interference. In the feature recognition layer, iris deflection interference was excluded by the improved methods of Gabor and Hamming and Haar and BP for the stable features extracted by the image processing layer, and two certification results were obtained by means of parallel recognition. The correct number of certifications for an algorithm under a certain lighting condition were counted. The method with the most correct number was set as the maximum trusted method under this lighting condition, and the results of the maximum trusted method were taken as the final decision, thereby eliminating the effect of illumination. Experiments using the JLU and CASIA iris libraries under the prerequisites in this paper show that the correct recognition rate of the algorithm can reach a high level of 98% or more, indicating that the algorithm can effectively improve the accuracy of the one-to-one certification of lightweight constrained state unsteady iris. Compared with the latest architecture algorithms, such as CNN and deep learning, the proposed algorithm is more suitable for the prerequisites presented in this paper, which has good environmental inclusiveness and can better improve existing traditional algorithms’ effectiveness through the design of a parallel integration model structure.

A study on the impact of data sampling rates on load signature event detection

The analysis of electrical load signatures is an enabling technology for many applications, such as ambient assisted living or energy-saving recommendations. Through the digitalization of electricity metering infrastructure, meter reading intervals are gradually becoming more frequent than the traditional once-per-year reporting. In fact, across smart meter generations, samples were initially reported in 15-min intervals, more recently once per second, and even newer devices capture readings at rates on the order of several kilohertz. The advantages of using such high sampling rates have, however, not been unambiguously demonstrated in literature. We thus choose a widely considered application scenario of energy data analytics, event detection, and assess the impact of the sampling rate choice on the correct event recognition rate. More specifically, we compare the accuracy of two event detection algorithms with respect to the resolution of their input data. The results of our analysis hint at a non-linear relation between accuracy and data resolution, yet also indicate that most event occurrences can be correctly determined when using a sampling rate of approximately 1 kHz, with only minimal improvements achievable through higher rates.

How Well do Orthopedic Surgeons Recognize Different Models of Total Elbow Arthroplasties on Plain Radiographs with the Use of a Diagnostic Flowchart?

Recognition of total elbow arthroplasties (TEAs) on plain radiographs is difficult due to a multitude of different types and models. Especially if surgery reports and documentation are not available, lost or when the primary surgery was performed in another hospital the prosthesis type may be undeterminable. Therefore we investigated in this platform study if a flowchart aids in recognition of thirteen different total elbow arthroplasty models on plain radiographs. An online questionnaire on the Shoulder and Elbow Platform was developed. Plain radiographs of thirteen TEA models were shown with and without the help of an especially developed flowchart describing distinguishing features. Ten orthopedic surgeons specialized in upper extremity surgery completed the study. Recognition rates of the thirteen total elbow arthroplasty models ranged between 20 and 100 percent without the flowchart. Using the flowchart recognition varied between 40 and 90 percent. The recognition rates with the flowchart were not significantly higher. Inter-observer reliability did not increase on a significant level. Correct recognition of total elbow arthroplasty models with plain radiographs remains imperfect with our developed flowchart. The flowchart increased correct recognition rates and inter-observer reliability.

Non-contact Gesture Recognition Technology Based on Low Frequency Electric Field

At the scene of natural disasters, it is safer and more convenient to remotely control robot completing search and rescue mission by gestures, but currently robot gesture control systems are difficult to maintain high efficiency in complex environmental conditions. Therefore, the paper proposes a non-contact detection system based on low frequency electric field, which can recognize various gestures. The system is not only not susceptible to the surrounding environment (light, sound), but also low cost and low power consumption. Based on the principle of electric field detection, we design the system structure, and perform simulation analysis. Followed, according to the simulation results, the signal processing circuit is designed to collect the detection electrode signals under different gestures. gesture features are extracted from a set of data, and random forest algorithm is used to recognize gestures. The results show that the detection system has a correct recognition rate of more than 90% and a response speed of about 3ms, which satisfies the application requirements.

Human action recognition based on HOIRM feature fusion and AP clustering BOW

In this paper, we propose a human action recognition method using HOIRM (histogram of oriented interest region motion) feature fusion and a BOW (bag of words) model based on AP (affinity propagation) clustering. First, a HOIRM feature extraction method based on spatiotemporal interest points ROI is proposed. HOIRM can be regarded as a middle-level feature between local and global features. Then, HOIRM is fused with 3D HOG and 3D HOF local features using a cumulative histogram. The method further improves the robustness of local features to camera view angle and distance variations in complex scenes, which in turn improves the correct rate of action recognition. Finally, a BOW model based on AP clustering is proposed and applied to action classification. It obtains the appropriate visual dictionary capacity and achieves better clustering effect for the joint description of a variety of features. The experimental results demonstrate that by using the fused features with the proposed BOW model, the average recognition rate is 95.75% in the KTH database, and 88.25% in the UCF database, which are both higher than those by using only 3D HOG+3D HOF or HOIRM features. Moreover, the average recognition rate achieved by the proposed method in the two databases is higher than that obtained by other methods.

Improved Damage Localization and Quantification of CFRP Using Lamb Waves and Convolution Neural Network

A novel method is proposed in this paper for simultaneously locating and quantifying damage in composite plates by employing Lamb waves and the algorithm of convolution neural network. The interaction between Lamb wave and damage of different degrees is also studied by simulation. The experiments on Lamb wave are carried out by employing a square array which is composed of four piezoelectric wafers. First of all, the sensor array collects response signals of Lamb wave as training data, and then de-noises them adopting the method of the wavelet transform. In the process, the damage caused to the composite can be realized through mass blocks. Besides, the Fourier transform is applied for the extraction of the characteristics shown by the signals. After that, the spectrum with the characteristics of damage and corresponding damage modes are employed as input and output of the convolutional neural network, respectively, and accordingly, the model of damage identification is established. Finally, 191 samples (from a total of 192) were identified accurately and the correct recognition rate achieved is 99.5%, which consequently demonstrates that the convolution neural network can be employed to establish the complex mapping relationship between signal and damage, and further proves that the proposed method performs well in high accuracy and great potential in simultaneous localization and quantitative identification of damage existing in composite plate.

Design and Development of Inventory System Based on Barcode Scanning Technology

The inventory machine is a device for managing the use of goods. In order to achieve the high efficiency and accuracy of data collection, processing and transmission of goods on site, a system of inventory machine based on barcode scanning technology is designed and developed. The system uses Java as programming language and designs a reasonable barcode scanning recognition mechanism. The external interference factors affecting barcode recognition and the problem that dirty barcode cannot be recognized are studied. Digital image processing technology is added to the inventory machine system, which improves the scanning speed of the inventory machine and the correct rate of barcode recognition. Through on-site testing, the system being developed meets the requirements of the enterprise for the inventory machine, and can complete the inventory and settlement of goods information efficiently and accurately in a fixed time. Compared with the previous use of the inventory machine, the cost and manpower cost in barcode scanning time are greatly reduced.

Research of OpenMV Intelligent Monitoring and Disease Identification System

Aiming at the requirement of video surveillance and disease identification in modern greenhouse, this paper designed a system consisted by greenhouse intelligent monitoring and disease identification system. At first, this paper using OpenMV camera designed a disease monitoring system which could recognize and collect disease images automatically by multi-color thresholds tracking. Then, this paper designed a cuckoo search and BP neural network collaborative search (CSBP-CS) algorithm, the algorithm combined the global search capability of Cuckoo Search(CS) and back-propagation algorithm of BP algorithm to optimize weights and thresholds collaboratively. this paper took three tomato diseases and normal leaves as research objects, firstly step was to separate the disease spots from disease images, then was to extract 56 classification features and select 47 excellent classification features by relief F to construct CSBP-CS RF classifier. Finally, this paper compared the classification accuracy of CSBP-CS with CSBP-CS RF network and analyzed the effectiveness of relief F. The simulation results showed that the average correct recognition rate of the CSBP-CS RF was approximately equal to CSBP-CS under the same conditions, but CSBP-CS RF is more simple, so Relief F can help to speed up the efficiency of CSBP-CS.

An experimental study for the effects of noise on face recognition algorithms under varying illumination

When the illumination changes, the appearance of facial images will change dramatically. Lighting changes make face recognition a very challenging and difficult job. In addition, the effects of noise on existing face recognition methods have been neglected in the literature, to the best of our knowledge. In this work, we study the effects of noise on existing illumination-invariant face recognition methods. We tested such noise as Gaussian white noise, Poisson noise, salt & pepper noise, speckle noise, etc. In total, 21 methods have been included in this study in this work. We find out that, when noise is added to facial images, Tan and Triggs’ method achieves the best results for both the extended Yale B face database and the CMU-PIE face database. When facial images do not contain noise, isotropic smoothing is preferred because it obtains the highest average recognition rate (96%) for the extended Yale B face database and 16 methods obtain perfect correct recognition rates (100%) for the CMU-PIE face database.

Adapted deep convnets technology for robust iris recognition

Iris recognition (IR) is widely adopted for human identification in the biometric fields. However, results using handcrafted features designed with the domain knowledge are obtained by conventional IR algorithms which are complex. Convolutional neural networks (convnets) have been proved to be robust with superior learning capability in many computer vision fields. The deep convnets perform better than shallower convnets. However, training the deep convnets for IR is difficult. IrisConvDeeper, an adaptive architecture based on convnets combining Softmax classifier that is used to alleviate the problems of training deep convnets for IR, is proposed. In addition, a better performance is obtained. Furthermore, a shallower architecture termed as IrisConvShallower is designed for comparison. IrisConvDeeper consists of structures called dense block and has 12 or 14 convolutional layers. Meanwhile, IrisConvShallower consists of six or seven standard convolutional layers. The performance of the proposed architectures is tested on two public iris databases, which are collected under different conditions: CASIA-Iris-V3 and IITD iris image databases. The experimental results demonstrate that the proposed IrisConvDeeper outperforms most of state-of-the-art approaches in terms of correct recognition rate, except some approaches that adopted specific methods and identified fewer subjects compared with our methods.

Radar Emitter Signal Identification Based on Multi-scale Information Entropy

With the increasing complexity of radar signals, it is more and more difficult to extract features of the real sequences, but when they are transformed to a symbol sequence, it is usually easier to mine the effective feature parameters. Therefore, a radar signal recognition method based on Multi-Scale Information Entropy (MSIE) is proposed. Firstly, the radar signal is transformed into symbolic sequence by Symbolic Aggregate approXimation (SAX) algorithm under different character number scales. Then, the information entropy of each symbol sequence is combined to form the MSIE feature vector. Finally, the k-Nearest Neighbor (k-NN) is used as a classifier to realize the classification and identification of radar signals. The simulation results of 6 typical radar signals show that using the proposed method the correct recognition rate of different radar signals is greater than 90% when Signal to Noise Ratio (SNR) is 5 dB, and better performance can be obtaned conpared with the traditional identification method based on complexity characteristics (box-dimension and sparseness).

Multimodal biometric system for ECG, ear and iris recognition based on local descriptors

Combination of multiple information extracted from different biometric modalities in multimodal biometric recognition system aims to solve the different drawbacks encountered in a unimodal biometric system. Fusion of many biometrics has proposed such as face, fingerprint, iris…etc. Recently, electrocardiograms (ECG) have been used as a new biometric technology in unimodal and multimodal biometric recognition system. ECG provides inherent the characteristic of liveness of a person, making it hard to spoof compared to other biometric techniques. Ear biometrics present a rich and stable source of information over an acceptable period of human life. Iris biometrics have been embedded with different biometric modalities such as fingerprint, face and palm print, because of their higher accuracy and reliability. In this paper, a new multimodal biometric system based ECG-ear-iris biometrics at feature level is proposed. Preprocessing techniques including normalization and segmentation are applied to ECG, ear and iris biometrics. Then, Local texture descriptors, namely 1D-LBP (One D-Local Binary Patterns), Shifted-1D-LBP and 1D-MR-LBP (Multi-Resolution) are used to extract the important features from the ECG signal and convert the ear and iris images to a 1D signals. KNN and RBF are used for matching to classify an unknown user into the genuine or impostor. The developed system is validated using the benchmark ID-ECG and USTB1, USTB2 and AMI ear and CASIA v1 iris databases. The experimental results demonstrate that the proposed approach outperforms unimodal biometric system. A Correct Recognition Rate (CRR) of 100% is achieved with an Equal Error Rate (EER) of 0.5%.

Nondestructive Visual Inspection Method of Double-Yolked Duck Egg

Duck eggs are rich in good protein and deeply favored by consumers. Among them, double-yolked duck eggs have high commercial value but are not easy to incubate, so they are often picked out by inspection before processing, circulation or being incubated in egg products industry. This paper used the computer vision technology on single- and double-yolked duck eggs with similar shape and size that were hard to distinguish merely by their external characteristics, studied the effective image processing algorithm and built a cognition model. Then, we collected duck eggs’ transmission light color image, extracted B channel image, carried out iterative morphological open reconstruction and threshold segmentation to acquire yolk area, used convexity defects to identify double-yolked eggs and finally utilized watershed and ellipse fitting to separate yolk. Through inspection, 150 single- and double-yolked eggs were distinguished by using this method, with correct recognition rate reaching to 99% and 100%, respectively. The experimental results indicate that this method has a high accuracy in recognizing double-yolked duck eggs and can provide technical support in nondestructive identification of double-yolked duck egg.