Image Recognition Approach Research Articles

Rotor fault diagnosis using a convolutional neural network with symmetrized dot pattern images

Vibration failure is a common problem in most rotating machinery, and vibration fault diagnosis is an important means of ensuring stable equipment operation. The present work proposes a rotor vibration fault diagnosis approach that transforms multiple vibration signals into symmetrized dot pattern (SDP) images, and then identifies the SDP graphical feature characteristic of different vibration states using a convolutional neural network (CNN). SDP images reveal different vibration states in a simple and intuitive manner. In addition, a CNN can reliably and accurately identify vibration faults by extracting the feature information of SDP images adaptively through deep learning. The proposed approach is tested experimentally using a rotor vibration test bed, and the results obtained are compared to those obtained with an equivalent CNN-based image recognition approach using orbit plot images. The rotor fault diagnosis precision is improved from 92% to 96.5%.

Development of framework for detecting smoking scene in video clips

<p>According to Global Adult Tobacco Survey 2016-17, 61.9% of people quitting tobacco the reason was the warnings displayed on the product covers. The focus of this paper is to automatically display warning messages in video clips. This paper explains the development of a system to automatically detect the smoking scenes using image recognition approach in video clips and then add the warning message to the viewer. The approach aims to detect the cigarette object using Tensorflow’s object detection API. Tensorflow is an open source software library for machine learning provided by Google which is broadly used in the field image recognition. At present, Faster R-CNN with Inception ResNet is theTensorflow’s slowest but most accurate model. Faster R-CNN with Inception Resnet v2 model is used to detect smoking scenes by training the model with cigarette as an object.</p><p><em><br /></em></p>

Deep-learned Top Tagging with a Lorentz Layer

We introduce a new and highly efficient tagger for hadronically decaying top quarks, based on a deep neural network working with Lorentz vectors and the Minkowski metric. With its novel machine learning setup and architecture it allows us to identify boosted top quarks not only from calorimeter towers, but also including tracking information. We show how the performance of our tagger compares with QCD-inspired and image-recognition approaches and find that it significantly increases the performance for strongly boosted top quarks.

Novel Image PreprocessingApproach for Automatic Speech Recognition

This research is intending to provide a novel approach of manipulating automatic speech recognition using image recognition approach. This research introduces hybrid 2D-Image-Hidden Markov Model(2DI)-(HMM) approach to handle preprocessing classification task in Automatic Speech Recognition System (ASR). The focus in this research is in the classification task. Due to that the proposed approach is novel and is a task in the whole ASR, it is evaluated using relative comparison to other popular approaches to run the same task on the same database. The relative comparison with hybrid Gaussian Mixture (GMM)-HMM with Mel Frequency Cepstral (MFCC) features is considered as reference results. This research introduces a new method of mapping speech signal into two-dimensionalspace. Speech stream is segmented and then the frequency contents are projected into frequency domain using a balanced tree structure filter. The wavelet packets technique is used to implement the filtering. The tree structure is captured into image. Database is constructed of encoded images. The imagesthenare segregated into speech classes. Hybrid Discrete Cosine Transform (DCT) based featuresare used for image encoding with (HMM) as Class model is evaluated against MFCC-HMM for the same classification problem. The proposed hybrid model indicates better balanced results over MFCC-HMM for handling the different classes. The considered classes in this research are vowels, consonants, plosives and speech silence.KED-TIMITCorpus is used in this research as source of speech information. This approach is indicating promising results especiallyin Silence and vowels detection.

An Entropy-Histogram Approach for Image Similarity and Face Recognition

Image similarity and image recognition are modern and rapidly growing technologies because of their wide use in the field of digital image processing. It is possible to recognize the face image of a specific person by finding the similarity between the images of the same person face and this is what we will address in detail in this paper. In this paper, we designed two new measures for image similarity and image recognition simultaneously. The proposed measures are based mainly on a combination of information theory and joint histogram. Information theory has a high capability to predict the relationship between image intensity values. The joint histogram is based mainly on selecting a set of local pixel features to construct a multidimensional histogram. The proposed approach incorporates the concepts of entropy and a modified 1D version of the 2D joint histogram of the two images under test. Two entropy measures were considered, Shannon and Renyi, giving a rise to two joint histogram-based, information-theoretic similarity measures: SHS and RSM. The proposed methods have been tested against powerful Zernike-moments approach with Euclidean and Minkowski distance metrics for image recognition and well-known statistical approaches for image similarity such as structural similarity index measure (SSIM), feature similarity index measure (FSIM) and feature-based structural measure (FSM). A comparison with a recent information-theoretic measure (ISSIM) has also been considered. A measure of recognition confidence is introduced in this work based on similarity distance between the best match and the second-best match in the face database during the face recognition process. Simulation results using AT&T and FEI face databases show that the proposed approaches outperform existing image recognition methods in terms of recognition confidence. TID2008 and IVC image databases show that SHS and RSM outperform existing similarity methods in terms of similarity confidence.

Abstract GS5-04: Prediction of occult invasive disease in ductal carcinoma in situ using deep learning features

Abstract Background: Deep learning, especially deep convolutional neural network (CNN), has emerged as a promising approach for many image recognition or classification tasks, demonstrating human or even superhuman performance. Used as feature extractor, some pre-trained CNN models can match or surpass the performance of domain-specific, “handcrafted” features. In this study, we aim to determine whether deep features extracted from digital mammograms using a pre-trained deep CNN are prognostic of occult invasive disease for patients with ductal carcinoma in situ (DCIS) on core needle biopsy. Materials and Methods: In this retrospective study, we collected digital mammography magnification views for 99 subjects with DCIS at biopsy, 25 of which were subsequently upstaged to invasive cancer. We utilized a deep CNN model that was pre-trained on non-medical images (e.g., animals, plants, instruments) as the feature extractor. Through a statistical pooling strategy, we extracted deep features at different levels of convolutional layers from the lesion areas, without sacrificing the original resolution or distorting the underlying topology. A multivariate classifier was then trained to predict which tumors contain occult invasive disease. This was compared to the performance of traditional “handcrafted” computer vision (CV) features previously developed specifically to assess mammographic calcifications. The generalization performance was assessed using Monte Carlo cross validation and receiver operating characteristic (ROC) curve analysis. Results: Deep features were able to distinguish DCIS with occult invasion from pure DCIS, with an area under the ROC curve (AUC-ROC) equal to 0.70 (95% CI: 0.68-0.73). This performance was comparable to the "handcrafted" CV features (AUC-ROC = 0.68, 95% CI: 0.66-0.71) that were designed with prior domain knowledge. Conclusion: In spite of being pre-trained on only non-medical images, the deep features extracted from digital mammograms demonstrated comparable performance to "handcrafted" CV features for the challenging task of predicting DCIS upstaging. Acknowledgments: This work was supported in part by NIH/NCI R01-CQA185138 and DOD Breast Cancer Research Program W81XWH-14-1-0473. Citation Format: Lo JY, Grimm LJ, Mazurowski MA, Baker JA, Marks JR, King LM, Maley CC, Hwang E-SS, Shi B. Prediction of occult invasive disease in ductal carcinoma in situ using deep learning features [abstract]. In: Proceedings of the 2017 San Antonio Breast Cancer Symposium; 2017 Dec 5-9; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2018;78(4 Suppl):Abstract nr GS5-04.

Mid‐level deep Food Part mining for food image recognition

There has been a growing interest in food image recognition for a wide range of applications. Among existing methods, mid‐level image part‐based approaches show promising performances due to their suitability for modelling deformable food parts (FPs). However, the achievable accuracy is limited by the FP representations based on low‐level features. Benefiting from the capacity to learn powerful features with labelled data, deep learning approaches achieved state‐of‐the‐art performances in several food image recognition problems. Both mid‐level‐based approaches and deep convolutional neural networks (DCNNs) approaches clearly have their respective advantages, but perhaps most importantly these two approaches can be considered complementary. As such, the authors propose a novel framework to better utilise DCNN features for food images by jointly exploring the advantages of both the mid‐level‐based approaches and the DCNN approaches. Furthermore, they tackle the challenge of training a DCNN model with the unlabelled mid‐level parts data. They accomplish this by designing a clustering‐based FP label mining scheme to generate part‐level labels from unlabelled data. They test on three benchmark food image datasets, and the numerical results demonstrate that the proposed approach achieves competitive performance when compared with existing food image recognition approaches.

Exposure-fusion-based dot-grid image acquisition and recognition for sheet metal strain analysis.

Dot-grid images are usually captured for grid strain analysis during sheet metal forming. Due to the strong reflective characteristic of the metallic surfaces, the recorded dot-grid images often have poor quality, low positioning accuracy, and low recognition rate. Therefore, an exposure-fusion-based dot-grid image acquisition and recognition approach is proposed. First, multiple dot-grid images are captured at different exposure levels. Subsequently, the recorded multi-exposure dot-grid images are fused into a new high-quality dot-grid image based on exposure fusion technology. Finally, a dot-grid image recognition procedure is developed to detect the dot-grids in the new dot-grid image. Both synthetic and real dot-grid images were tested to verify the performance of the novel approach. When synthetic dot-grid images were tested, the maximum positioning error was up to 6.044 pixels if they were recognized in the traditional way, whereas the maximum positioning error was reduced to 0.132 pixels if the novel approach was adopted. When real dot-grid images were tested, the lowest recognition rate is only 50.52% if they were recognized in the traditional way. Nevertheless, the recognition rate can reach about 91% if the novel approach was employed. These experimental results show the superiorities of the novel approach.

Deep Learning for Image-Based Cassava Disease Detection.

Cassava is the third largest source of carbohydrates for human food in the world but is vulnerable to virus diseases, which threaten to destabilize food security in sub-Saharan Africa. Novel methods of cassava disease detection are needed to support improved control which will prevent this crisis. Image recognition offers both a cost effective and scalable technology for disease detection. New deep learning models offer an avenue for this technology to be easily deployed on mobile devices. Using a dataset of cassava disease images taken in the field in Tanzania, we applied transfer learning to train a deep convolutional neural network to identify three diseases and two types of pest damage (or lack thereof). The best trained model accuracies were 98% for brown leaf spot (BLS), 96% for red mite damage (RMD), 95% for green mite damage (GMD), 98% for cassava brown streak disease (CBSD), and 96% for cassava mosaic disease (CMD). The best model achieved an overall accuracy of 93% for data not used in the training process. Our results show that the transfer learning approach for image recognition of field images offers a fast, affordable, and easily deployable strategy for digital plant disease detection.

RETRACTED ARTICLE: Image pattern recognition in big data: taxonomy and open challenges: survey

Image pattern recognition in the field of big data has gained increasing importance and attention from researchers and practitioners in many domains of science and technology. This paper focuses on the usage of image pattern recognition for big data applications. In this context, the taxonomy of image pattern recognition and big data is revealed. The applications of image pattern recognition for big data, including multimedia, biometrics, and biology/biomedical, are also highlighted. Moreover, the significance of using pattern-based feature reduction in big data is discussed, and machine-learning techniques in pattern recognition applications are presented. A comparison based on the objectives of the approaches is presented to underline the taxonomy. This paper provides a novel review in exploring image recognition approaches for big data, which can be used in future research.

Assessment of rooftop photovoltaic potentials at the urban level using publicly available geodata and image recognition techniques

Abstract The local generation of renewable electricity through roof-mounted photovoltaic (PV) systems on buildings in urban areas provides huge potentials for the mitigation of greenhouse gas emissions. This contribution presents a new method to provide local decision makers with tools to assess the remaining PV potential within their respective communities. It allows highly detailed analyses without having to rely on 3D city models, which are often not available. This is achieved by a combination of publicly available geographical building data and aerial images that are analyzed using image recognition and machine learning approaches. The method also employs sophisticated algorithms for irradiance simulation and power generation that exhibit a higher accuracy than most existing PV potential studies. The method is demonstrated with an application to the city of Freiburg, for which a technical PV electricity generation potential of about 524 GWh/a is identified. A validation with a 3D city model shows that the correct roof azimuth can be determined with an accuracy of about 70% and existing solar installations can be detected with an accuracy of about 90%. This demonstrates that the method can be employed for spatially and temporally detailed PV potential assessments in arbitrary urban areas when only public geographical building data is available instead of exact 3D city model data. Future work will focus on methodological improvements as well as on the integration of the method within an urban energy system modeling framework.

An effective two-dimensional linear discriminant analysis with locality preserving approach for image recognition

Linear discriminant analysis (LDA) is one of the most popular feature extraction methods in image recognition. Nevertheless, LDA always suffers from some undesired behaviors caused by globality, namely ignoring local geometric structure of images. To fully capture the geometric information, we propose a novel two-dimensional linear discriminant analysis with locality preserving (2DLP-LDA) approach, which can cover both within-class and between-class local geometric information. 2DLP-LDA re-characterizes the within-class scatter matrix by employing locality preserving projection technique and re-characterizes the between-class scatter matrix by introducing a Gaussian weighting function, making the samples from the same class moderately cluster, while the samples from different classes properly distribute in the reduced subspace. Encouraging experimental results on FERET face database as well as SEU bovine iris database show the effectiveness of the proposed approach.

Development and validation of a noninvasive 2-gene molecular assay for cutaneous melanoma

BackgroundClinical and histopathologic assessment of pigmented skin lesions remains challenging even for experts. Differentiated and accurate noninvasive diagnostic modalities are highly desirable.ObjectiveWe sought to provide clinicians with such a tool.MethodsA 2-gene classification method based on LINC00518 and preferentially expressed antigen in melanoma (PRAME) gene expression was evaluated and validated in 555 pigmented lesions (157 training and 398 validation samples) obtained noninvasively via adhesive patch biopsy. Results were compared with standard histopathologic assessment in lesions with a consensus diagnosis among 3 experienced dermatopathologists.ResultsIn 398 validation samples (87 melanomas and 311 nonmelanomas), LINC00518 and/or PRAME detection appropriately differentiated melanoma from nonmelanoma samples with a sensitivity of 91% and a specificity of 69%. We established LINC00518 and PRAME in both adhesive patch melanoma samples and underlying formalin fixed paraffin embedded (FFPE) samples of surgically excised primary melanomas and in melanoma lymph node metastases.LimitationsThis technology cannot be used on mucous membranes, palms of hands, and soles of feet.ConclusionsThis noninvasive 2-gene pigmented lesion assay classifies pigmented lesions into melanoma and nonmelanoma groups and may serve as a tool to help with diagnostic challenges that may be inherently linked to the visual image and pattern recognition approach.

Estimation of rooftop solar photovoltaic potential using geo-spatial techniques: A perspective from planned neighborhood of Karachi – Pakistan

With an ever increasing population and significant solar resource availability throughout the whole year, Karachi metropolis hold a promising rooftop solar photovoltaic (PV) potential considering its millions of urban households. This research work highlights techniques to combine geographic information systems (GIS) and object-based image recognition approach to identify the available rooftop area for PV deployment in a small scale region of DHA Phase 7 Karachi. A six step methodology has been adopted for the estimation of rooftop PV potential which involves geographic division of high resolution satellite imagery; sampling and rooftop feature extraction using FE tool of ENVI EX software; extrapolation of rooftop areas for the entire ROI using roof area-population relationship; visual inspection to analyze different rooftop factors such as building orientation, shading effect from trees and nearby buildings and other roof uses; a comparison of extracted rooftops to the physically measured sample rooftops, reduction for shading and other uses; and finally conversion to energy and power outputs. A relationship of total roof area and population of 13 m2/capita ±5% has been found. With higher efficiency rooftop PV panels, 12.24 MW of solar power can be generated which is 122.4% of peak power demand of DHA Phase 7.

Comparison of Feed Forward Network and Radial Basis Function for Detecting and Recognition of License Plate

One of the most important topics of intelligent transportation system is the License Plate Recognition (LPR). LPR systems have many potential applications in intelligent traffic systems, such as the payment of parking fee, highway toll fee, traffic data collection, traffic monitoring systems, traffic law enforcement, security control of restricted areas and so on. LPR was developed to identify vehicles by the contents of their license plates. The LPR system consists of three major modules: license plate extraction, segmentation and recognition of individual characters. This paper presents a study of applying the neural network approach for character image recognition. The new approach is tested on 400 samples of extracted license plate images captured in outdoor environment. The result yield 99.2% recognition accuracy, the method takes 1.6 seconds to perform the car plate recognition from vehicle's image. In order to decrease problems such as low quality and low contrast in the vehicle images, image recognition is done by the two different methods first is feed forward method and another is Radial Basis function and the best one is selected. The algorithm based on neural network which can quickly and correctly detect the region of license plate and the license plate detecting rate of success is 99.2%.

A Novel Approach for Image Recognition to Enhance the Quality of Decision Making by Applying Degree of Correlation Using Artificial Neural Networks

Many diversified applications do exist in science & technology, which make use of the primary theory of a recognition phenomenon as one of its solutions. Recognition scenario is incorporated with a set of decisions and the action according to the decision purely relies on the quality of extracted information on utmost applications. Thus, the quality decision making absolutely reckons on processing momentum and precision which are entirely coupled with recognition methodology. In this article, a latest rule is formulated based on the degree of correlation to characterize the generalized recognition constraint and the application is explored with respect to image based information extraction. Machine learning based perception called feed forward architecture of Artificial Neural Network has been applied to attain the expected eminence of elucidation. The proposed method furnishes extraordinary advantages such as less memory requirements, extremely high level security for storing data, exceptional speed and gentle implementation approach.

Recognition of Low-Resolution Logos in Vehicle Images Based on Statistical Random Sparse Distribution

Traditional image recognition approaches can achieve high performance only when the images have high resolution and superior quality. A new vehicle logo recognition (VLR) method is proposed to treat low-resolution and poor-quality images captured from urban crossings in intelligent transport system, and the proposed approach is based on statistical random sparse distribution (SRSD) feature and multiscale scanning. The SRSD feature is a novel feature representation strategy that uses the correlation between random sparsely sampled pixel pairs as an image feature and describes the distribution of a grayscale image statistically. Multiscale scanning is a creative classification algorithm that locates and classifies a logo integrally, which alleviates the effect of propagation errors in traditional methods by processing the location and classification separately. Experiments show an overall recognition rate of 97.21% for a set of 3370 vehicle images, which showed that the proposed algorithm outperforms classical VLR methods for low-resolution and inferior quality images and is very suitable for on-site supervision in ITSs.

FUZZY BASED IMAGE DIMENSIONALITY REDUCTION USING SHAPE PRIMITIVES FOR EFFICIENT FACE RECOGNITION

Today face recognition capability of the human visual system plays a significant role in day to day life due to numerous important applications for automatic face recognition. One of the problems with the recent image classification and recognition approaches are they have to extract features on the entire image and on the large grey level range of the image. The present paper overcomes this by deriving an approach that reduces the dimensionality of the image using Shape primitives and reducing the grey level range by using a fuzzy logic while preserving the significant attributes of the texture. The present paper proposed an Image Dimensionality Reduction using shape Primitives (IDRSP) model for efficient face recognition. Fuzzy logic is applied on IDRSP facial model to reduce the grey level range from 0 to 4. This makes the proposed fuzzy based IDRSP (FIDRSP) model suitable to Grey level co-occurrence matrices. The proposed FIDRSP model with GLCM features are compared with existing face recognition algorithm. The results indicate the efficacy of the proposed method.

Kernel Feature Extraction Approach for Color Image Recognition

Color Image Recognition is one of the most important fields in Pattern Recognition. Both Multi-set canonical correlation analysis and Kernel method are important techniques in the field of color image recognition. In this paper, we combine the two methods and propose one novel color image recognition approach: color image kernel canonical correlation analysis (CIKCCA). Color image kernel canonical correlation analysis is based on the theory of multi-set canonical correlation analysis and extracts canonical correlation features among the color image components. Then fuse the features of the color image components in the feature level, which are used for classification and recognition. Experimental results on the FRGC-v2 public color image databases demonstrate that the proposed approach acquire better recognition performance than other color recognition methods.

Sélection adaptative de caractéristiques pertinentes et classification hiérarchique des images dans les bases hétérogènes

In heterogeneous databases, images often provided from different sources and belong to different topics, hence there is a need for a large description to ensure efficient representation of their content. However, extracted features are not always adapted to the considered image database. In this paper we propose a new image recognition approach based on two innovations, namely adaptive feature selection and Multi-Model Classification Method (MC-MM). The adaptive selection considers only the most adapted features with the used image database content. The MC-MM method ensures image recognition using hierarchically selected features. Experimental results confirm the effectiveness and the robustness of our proposed approach.