Bag Of Features Research Articles

Improving cloud type classification of ground-based images using region covariance descriptors

Abstract. The distribution and frequency of occurrence of different cloud types affect the energy balance of the Earth. Automatic cloud type classification of images continuously observed by the ground-based imagers could help climate researchers find the relationship between cloud type variations and climate change. However, by far it is still a huge challenge to design a powerful discriminative classifier for cloud categorization. To tackle this difficulty, in this paper, we present an improved method with region covariance descriptors (RCovDs) and the Riemannian bag-of-feature (BoF) method. RCovDs model the correlations among different dimensional features, which allows for a more discriminative representation. BoF is extended from Euclidean space to Riemannian manifold by k-means clustering, in which Stein divergence is adopted as a similarity metric. The histogram feature is extracted by encoding RCovDs of the cloud image blocks with a BoF-based codebook. The multiclass support vector machine (SVM) is utilized for the recognition of cloud types. The experiments on the ground-based cloud image datasets show that a very high prediction accuracy (more than 98 % on two datasets) can be obtained with a small number of training samples, which validate the proposed method and exhibit the competitive performance against state-of-the-art methods.

Stable bagging feature selection on medical data

In the medical field, distinguishing genes that are relevant to a specific disease, let’s say colon cancer, is crucial to finding a cure and understanding its causes and subsequent complications. Usually, medical datasets are comprised of immensely complex dimensions with considerably small sample size. Thus, for domain experts, such as biologists, the task of identifying these genes have become a very challenging one, to say the least. Feature selection is a technique that aims to select these genes, or features in machine learning field with respect to the disease. However, learning from a medical dataset to identify relevant features suffers from the curse-of-dimensionality. Due to a large number of features with a small sample size, the selection usually returns a different subset each time a new sample is introduced into the dataset. This selection instability is intrinsically related to data variance. We assume that reducing data variance improves selection stability. In this paper, we propose an ensemble approach based on the bagging technique to improve feature selection stability in medical datasets via data variance reduction. We conducted an experiment using four microarray datasets each of which suffers from high dimensionality and relatively small sample size. On each dataset, we applied five well-known feature selection algorithms to select varying number of features. The proposed technique shows a significant improvement in selection stability while at least maintaining the classification accuracy. The stability improvement ranges from 20 to 50 percent in all cases. This implies that the likelihood of selecting the same features increased 20 to 50 percent more. This is accompanied with the increase of classification accuracy in most cases, which signifies the stated results of stability.

An Improved Content Based Image Retrieval Technique by Exploiting Bi-layer Concept

Applications for retrieving similar images from a large collection of images have increased significantly in various fields with the rapid advancement of digital communication technologies and exponential evolution in the usage of the Internet. Content-based image retrieval (CBIR) is a technique to find similar images on the basis of extracting the visual features such as color, texture, and/or shape from the images themselves. During the retrieval process, features and descriptors of the query image are compared to those of the images in the database to rank each indexed image accordingly to its distance to the query image. This paper has developed a new CBIR technique which entails two layers, called bi-layers. In the first layer, all images in the database are compared to the query image based on the bag of features (BoF) technique, and hence, the M most similar images to the query image are retrieved. In the second layer, the M images obtained from the first layer are compared to the query image based on the color, texture, and shape features to retrieve the N number of the most similar images to the query image. The proposed technique has been evaluated using a well-known dataset of images called Corel-1K. The obtained results revealed the impact of exploring the idea of bi-layers in improving the precision rate in comparison to the current state-of-the-art techniques in which achieved precision rate of 82.27% and 76.13% for top-10 and top-20, respectively.

Scene Semantic Recognition Based on Modified Fuzzy C-Mean and Maximum Entropy Using Object-to-Object Relations

With advances in machine vision systems (e.g., artificial eye, unmanned aerial vehicles, surveillance monitoring) scene semantic recognition (SSR) technology has attracted much attention due to its related applications such as autonomous driving, tourist navigation, intelligent traffic and remote aerial sensing. Although tremendous progress has been made in visual interpretation, several challenges remain (i.e., dynamic backgrounds, occlusion, lack of labeled data, changes in illumination, direction, and size). Therefore, we have proposed a novel SSR framework that intelligently segments the locations of objects, generates a novel Bag of Features, and recognizes scenes via Maximum Entropy. First, denoising and smoothing are applied on scene data. Second, modified Fuzzy C-Means integrates with super-pixels and Random Forest for the segmentation of objects. Third, these segmented objects are used to extract a novel Bag of Features that concatenate different blobs, multiple orientations, Fourier transform and geometrical points over the objects. An Artificial Neural Network recognizes the multiple objects using the different patterns of objects. Finally, labels are estimated via Maximum Entropy model. During experimental evaluation, our proposed system illustrated a remarkable mean accuracy rate of 90.07% over the MSRC dataset and 89.26% over the Caltech 101 for object recognition, and 93.53% over the Pascal-VOC12 dataset for scene recognition, respectively. The proposed system should be applicable to various emerging technologies, such as augmented reality, to represent the real-world environment for military training and engineering design, as well as for entertainment, artificial eyes for visually impaired people and traffic monitoring to avoid congestion or road accidents.

Deep Sparse Auto-Encoder Features Learning for Arabic Text Recognition

One of the most recent challenging issues of pattern recognition and artificial intelligence is Arabic text recognition. This research topic is still a pervasive and unaddressed research field, because of several factors. Complications arise due to the cursive nature of the Arabic writing, character similarities, unlimited vocabulary, use of multi-size and mixed-fonts, etc. To handle these challenges, an automatic Arabic text recognition requires building a robust system by computing discriminative features and applying a rigorous classifier together to achieve an improved performance. In this work, we introduce a new deep learning based system that recognizes Arabic text contained in images. We propose a novel hybrid network, combining a Bag-of-Feature (BoF) framework for feature extraction based on a deep Sparse Auto-Encoder (SAE), and Hidden Markov Models (HMMs), for sequence recognition. Our proposed system, termed BoF-deep SAE-HMM, is tested on four datasets, namely the printed Arabic line images Printed KHATT (P-KHATT), the benchmark printed word images Arabic Printed Text Image (APTI), the benchmark handwritten Arabic word images IFN/ENIT, and the benchmark handwritten digits images Modified National Institute of Standards and Technology (MNIST).

Optimal Metric Evaluation-Based Multicue Inverse Sparse Appearance Model for Object Tracking

In order to obtain the discriminative compact appearance model for tracking objects effectively, this paper proposes a new structural tracking strategy that includes multicue inverse sparse appearance model and optimal metric evaluation between online robust templates and a limited number of particle samples in the looping process. Multicue inverse sparse appearance model globally improves the efficient selection of informative particle samples that can avoid the cumbersome coding and decoding cost for the trivial random particle samples. Only the most potential crucial cases are involved in each tracking loop. This refrains from unreasonable, rough numerical reduction of particle samples and also keeps the unbiasedness and dynamic stochasticness of the sampling process. Meanwhile, low-rank self-representatives for positive and negative samples facilitate the formulation of a suitable code book that arranges the useful sparse coefficients for feature bags and facilitates optimal metric evaluation for online training. It also alleviates the accuracy degradation of tracking occluded objects and improves the robustness of the tracker. Both of them preserve the discriminative compactness of target which speeds up particle filtering localization to separate the target object from distractors. Moreover, the proposed method exploits online appearance representations to learn the sharing compact information that avoids massive calculation burdens for massive visual data.

Temporal Bag-of-Features Learning for Predicting Mid Price Movements Using High Frequency Limit Order Book Data

Time-series forecasting has various applications in a wide range of domains, e.g., forecasting stock markets using limit order book data. Limit order book data provide much richer information about the behavior of stocks than its price alone, but also bear several challenges, such as dealing with multiple price depths and processing very large amounts of data of high dimensionality, velocity, and variety. A well-known approach for efficiently handling large amounts of high-dimensional data is the bag-of-features (BoF) model. However, the BoF method was designed to handle multimedia data such as images. In this paper, a novel temporal-aware neural BoF model is proposed tailored to the needs of time-series forecasting using high frequency limit order book data. Two separate sets of radial basis function and accumulation layers are used in the temporal BoF to capture both the short-term behavior and the long-term dynamics of time series. This allows for modeling complex temporal phenomena that occur in time-series data and further increase the forecasting ability of the model. Any other neural layer, such as feature transformation layers, or classifiers, such as multilayer perceptrons, can be combined with the proposed deep learning approach, which can be trained end-to-end using the back-propagation algorithm. The effectiveness of the proposed method is validated using a large-scale limit order book dataset, containing over 4.5 million limit orders, and it is demonstrated that it greatly outperforms all the other evaluated methods.

An efficient method to classify GI tract images from WCE using visual words

The digital images made with the Wireless Capsule Endoscopy (WCE) from the patient's gastrointestinal tract are used to forecast abnormalities. The big amount of information from WCE pictures could take 2 hours to review GI tract illnesses per patient to research the digestive system and evaluate them. It is highly time consuming and increases healthcare costs considerably. In order to overcome this problem, the CS-LBP (Center Symmetric Local Binary Pattern) and the ACC (Auto Color Correlogram) were proposed to use a novel method based on a visual bag of features (VBOF). In order to solve this issue, we suggested a Visual Bag of Features(VBOF) method by incorporating Scale Invariant Feature Transform (SIFT), Center-Symmetric Local Binary Pattern (CS-LBP) and Auto Color Correlogram (ACC). This combination of features is able to detect the interest point, texture and color information in an image. Features for each image are calculated to create a descriptor with a large dimension. The proposed feature descriptors are clustered by K- means referred to as visual words, and the Support Vector Machine (SVM) method is used to automatically classify multiple disease abnormalities from the GI tract. Finally, post-processing scheme is applied to deal with final classification results i.e. validated the performance of multi-abnormal disease frame detection.

Smart Handheld Based Human Activity Recognition Using Multiple Instance Multiple Label Learning

Human activity recognition (HAR) and monitoring is beneficial for many medical applications, such as eldercare and post-trauma rehabilitation after surgery. HAR models based on smartphone’s accelerometer data could provide a convenient and ubiquitous solution to this problem. However, such models are mostly concerned with identifying basic activities such as ‘stand’/‘walk’ and thus the high-level context such as ‘walk in a queue’ for which a set of specific activities is performed remain unnoticed. Consequently, in this paper, we design a HAR framework that can identify a group of activities (rather than a single basic activity) being performed in a time window, thus, enables us to extract more meaningful information about the subject’s overall context. An algorithm is designed to formulate HAR as a multi-instance multi-label (MIML) learning problem. The procedure of generating feature bags of consecutive activity traces having multiple labels is formulated. In this work, the temporal relationship among activities is exploited to obtain a more comprehensive HAR model. Interestingly, the framework is found to completely/partially identify activity sequences that may not even be present in the training dataset. The framework is implemented and found to be working adequately when tested with real dataset collected from 8 users for 12 different activity combinations. MIML-kNN is found to provide maximum average precision (around 90%) even for an unseen test data-set.

Well-Known brands recognition by automated classifiers using local and global features

From color and type to patterns and illustrations, brands sense to be recognizable and convey their values and personality. Here patterns and color are key elements, as they can play a vital role in brand recognition. The images used for brand classification were handpicked and collectively named as HKDataset. We have explored various feature extractors used for classification and used automated classifiers named Linear SVM to achieve higher accuracy while tuning the model parameters to achieve optimal performance. It has been observed that Support Vector Machines performs better when using GIST descriptors combined with Bag of SIFT features. We hope to apply deep learning and other sophisticated classifiers to much-expanded categories of brands in the future.

Predicting TF-DNA Binding Motifs from ChIP-seq Datasets Using the Bag-Based Classifier Combined With a Multi-Fold Learning Scheme

The rapid development of high-throughput sequencing technology provides unique opportunities for studying of transcription factor binding sites, but also brings new computational challenges. Recently, a series of discriminative motif discovery (DMD) methods have been proposed and offer promising solutions for addressing these challenges. However, because of the huge computation cost, most of them have to choose approximate schemes that either sacrifice the accuracy of motif representation or tune motif parameter indirectly. In this paper, we propose a bag-based classifier combined with a multi-fold learning scheme (BCMF) to discover motifs from ChIP-seq datasets. First, BCMF formulates input sequences as a labeled bag naturally. Then, a bag-based classifier, combining with a bag feature extracting strategy, is applied to construct the objective function, and a multi-fold learning scheme is used to solve it. Compared with the existing DMD tools, BCMF features three improvements: 1) Learning position weight matrix (PWM) directly in a continuous space; 2) Proposing to represent a positive bag with a feature fused by its k "most positive" patterns. 3) Applying a more advanced learning scheme. The experimental results on 134 ChIP-seq datasets show that BCMF substantially outperforms existing DMD methods (including DREME, HOMER, XXmotif, motifRG, EDCOD and our previous work).

Image-based soft drink type classification and dietary assessment system using deep convolutional neural network with transfer learning

Nowadays, people are taking soft drinks (carbonated nonalcoholic beverages) at an increasing rate. Health experts around the world have cautioned from time to time that these drinks lead to weight gain, raise the risk of non-communicable diseases, and so on. To develop consciousness among people, the present work describes an image-based tool to self-monitor the nutritional information of soft drinks by using a deep convolutional neural network (CNN) along with transfer learning. At first, a pre-processing function is done through noise reduction and contrast enhancement. Then the location of the drinks region is extracted through visual saliency and mean-shift segmentation technique. After removing backgrounds and segment out only the region of interest from the image a deep CNN-based transfer learning model is employed for the drink classification. Finally, the size of each drink bottle is estimated using the bag-of-feature (BoF) and distance ratio calculation to find the nutrition value from the nutrition fact table. To perform experimentation a dataset is built containing ten most consumed soft drinks in Bangladesh using images from the ImageNet dataset, internet sources and also self-capturing. The experiment confirms that our system can detect and recognize different types of drinks with an accuracy of 98.51%.

Deep learning for non-rigid 3D shape classification based on informative images

In order to enhance the discernment of features in view-based 3D shape recognition, we propose a joint convolutional neural network (CNN) learning model based on informative images. It learns deep features from intrinsic feature images and extrinsic 2D views, and generates a synthetic feature vector via weighted aggregation and refinement process, which has achieved remarkable improvement in non-rigid 3D shape classification. Our joint CNNs model contains three parts: the first part is the geometry-based feature generation unit. We provide a discriminative BoF (bag of features) image descriptor and construct CNN framework to learn the geometric features of the model. The second part is the view-based feature generation unit. We establish a parallel CNN to extract spatial features from optimized 2D views. The third part is a score generation and refinement unit, which automatically learns the weighted scores of geometric features and spatial features. Finally, the aggregated feature is refined in a CNN framework and serves as an informative shape descriptor for recognition task. The experimental results demonstrate that our deep features have the strong discerning ability. Thus, better performance and robustness can be obtained compared to state-of-the-art methods.

Person Reidentification pada Sistem Surveillance Cerdas Menggunakan Metode Bag of Visual Feature

Sistem pengawasan berbasis CCTV telah banyak digunakan untuk memantau tempat umum, sehingga keamanan dan kenyamanan lebih terjamin. Salah satu permasalahan pada proses pengawasan adalah mengidentifikasi objek manusia pada kamera CCTV yang berbeda, yang disebut sebagai person reidentification (Re-ID). Re-ID adalah proses untuk mengidentifikasi gambar objek manusia, yang berasal dari dua citra atau lebih yang berada di sudut pandang kamera CCTV berbeda, adalah orang yang sama atau tidak. Pada makalah ini diusulkan metode berbasis fitur visual dari citra objek dengan nama Bag of Visual Feature (BOVF). BOVF bekerja dengan merepresentasikan data sebagai kumpulan fitur lokal yang digunakan dengan mekanisme pengelompokan fitur (feature clustering). Implementasi BOVF menggunakan algoritme pengelompokan fitur Density Based Spatial Clustering of Applications with Noise (DBSCAN) pada proses pelatihan model citra dengan mempelajari fitur Histogram of Oriented Gradient (HOG). Hasil dari penelitian yang menggunakan dataset sebanyak 70 citra frame dari dataset iLIDS-VID ini mendapatkan akurasi terbaik pada R-20 sebesar 88% menggunakan DBSCAN dengan kecepatan proses sebesar 1,85 detik.

Bag-of-features-based radiomics for differentiation of ocular adnexal lymphoma and idiopathic orbital inflammation from contrast-enhanced MRI.

To evaluate the effectiveness of bag-of-features (BOF)-based radiomics for differentiating ocular adnexal lymphoma (OAL) and idiopathic orbital inflammation (IOI) from contrast-enhanced MRI (CE-MRI). Fifty-six patients with pathologically confirmed IOI (28 patients) and OAL (28 patients) were randomly divided into training (n = 42) and testing (n = 14) groups. One hundred sixty texture features extracted from the CE-MR image were encoded into the BOF representation with fewer features. The support vector machine (SVM) with a linear kernel was used as the classifier. Data augmented was performed by cropping orbital lesions in different directions to alleviate the over-fitting problem. Student's t test and the Holm-Bonferroni method were employed to compare the performance of different analysis methods. The chi-square test was used to compare the analysis with MRI and human radiological diagnosis. In the independent testing group, the differentiation by the BOF features with augmentation achieved an area under the curve (AUC) of 0.803 (95% CI: 0.725-0.880), which was significantly higher than that of the BOF features without augmentation and that of the texture features (p < 0.05). In addition, the same radiomic analysis with pre-contrast MRI obtained an AUC of 0.618 (95% CI: 0.560-0.677), which was significantly lower than that with CE-MRI. The diagnostic performance of the analysis with CE-MRI was significantly better than the radiology resident (p < 0.05) but had no significant difference with the experienced radiologist, even though there was less consistency between the radiomic analysis and the human visual diagnosis. The BOF-based radiomics may be helpful for the differentiation between OAL and IOI. • It is challenging to differentiate OAL from IOI due to the similar clinical and image features. • Radiomics has great potential for the noninvasive diagnosis of orbital diseases. • The BOF representation from patch to image may help the differentiation of OAL and IOI.

Two-stream spatiotemporal feature fusion for human action recognition

Human action recognition is still a challenging topic in the computer vision field that has attracted a large number of researchers. It has a significant importance in varieties of applications such as intelligent video surveillance, sports analysis, and human–computer interaction. Recent works attempt to exploit the progress in deep learning architecture to learn spatial and temporal features from action video. However, it remains unclear how to combine spatial and temporal information with convolutional neural network. In this paper, we propose a novel human action recognition method by fusing spatial and temporal features learned from a simple unsupervised convolutional neural network called principal component analysis network (PCANet) in combination with bag-of-features (BoF) and vector of locally aggregated descriptors (VLAD ) encoding schemes. Firstly, both spatial and temporal features are learned via PCANet using a subset of frames and temporal templates for each video, while their dimensionality is reduced using whitening transformation (WT). The temporal templates are calculated using short-time motion energy images (ST-MEI) based on frame differencing. Then, the encoding scheme is applied to represent the final dual spatiotemporal PCANet features by feature fusion. Finally, the support vector machine (SVM) classifier is exploited for action recognition. Extensive experiments have been performed on two popular datasets, namely KTH and UCF sports, to evaluate the performance of proposed method. Our experimental results using leave-one-out evaluation strategy demonstrate that the proposed method presents satisfactory and comparable results on both datasets.

Estimation of Personalized Heterogeneous Treatment Effects Using Concatenation and Augmentation of Feature Vectors

A new meta-algorithm for estimating the conditional average treatment effects is pro-posed in the paper. The basic idea behind the algorithm is to consider a new dataset consisting of feature vectors produced by means of concatenation of examples from control and treatment groups, which are close to each other. Outcomes of new data are defined as the difference between outcomes of the corresponding examples comprising new feature vectors. The second idea is based on the assumption that the number of controls is rather large and the control outcome function is precisely determined. This assumption allows us to augment treatments by generating feature vectors which are closed to available treatments. The outcome regression function constructed on the augmented set of concatenated feature vectors can be viewed as an estimator of the conditional average treatment effects. A simple modification of the Co-learner based on the random subspace method or the feature bagging is also proposed. Various numerical simulation experiments illustrate the proposed algorithm and show its outperformance in comparison with the well-known T-learner and X-learner for several types of the control and treatment outcome functions.

Leaf Image Recognition Based on Bag of Features

Plants are ubiquitous in human life. Recognizing an unknown plant by its leaf image quickly is a very interesting and challenging research. With the development of image processing and pattern recognition, plant recognition based on image processing has become possible. Bag of features (BOF) is one of the most powerful models for classification, which has been used for many projects and studies. Dual-output pulse-coupled neural network (DPCNN) has shown a good ability for texture features in image processing such as image segmentation. In this paper, a method based on BOF and DPCNN (BOF_DP) is proposed for leaf classification. BOF_DP achieved satisfactory results in many leaf image datasets. As it is hard to get a satisfactory effect on the large dataset by a single feature, a method (BOF_SC) improved from bag of contour fragments is used for shape feature extraction. BOF_DP and LDA (linear discriminant analysis) algorithms are, respectively, employed for textual feature extraction and reducing the feature dimensionality. Finally, both features are used for classification by a linear support vector machine (SVM), and the proposed method obtained higher accuracy on several typical leaf datasets than existing methods.

Investigation of 3D surface profile reconstruction technology for automatic evaluation of fabric smoothness appearance

Objective evaluation of smoothness appearance of fabrics after repeated home laundering is not only the key point, but also a difficulty in the field of textile inspection and clothing industry. This paper reports a 3D reconstruction system and method for objective assessment of fabric smoothness appearance based on Multi-view Stereo Vision (MVS) and Bag-of-Features (BOF). An image acquisition system for texture-less fabric was designed with a specific illumination module. Three-dimensional surface profile of wrinkled fabric was reconstructed by Structure from Motion (SFM) and Patch-based Multi-view Stereo (PMVS) techniques. To facilitate feature extraction, the generated 3D point cloud was mapped to a 2D image plane. The Bag-of-Features (BOF) model was utilized to characterize the fabric wrinkles and the K-Nearest Neighbor (KNN) algorithm was used to achieve automatic evaluation of fabric smoothness appearance. The experimental results with a high recognition of 93.33% indicated the effectiveness and stability of the proposed method.

Bag of feature and support vector machine based early diagnosis of skin cancer

Skin cancer is one of the diseases which lead to death if not detected at an early stage. Computer-aided detection and diagnosis systems are designed for its early diagnosis which may prevent biopsy and use of dermoscopic tools. Numerous researches have considered this problem and achieved good results. In automatic diagnosis of skin cancer through computer-aided system, feature extraction and reduction plays an important role. The purpose of this research is to develop computer-aided detection and diagnosis systems for classifying a lesion into cancer or non-cancer owing to the usage of precise feature extraction technique. This paper proposed the fusion of bag-of-feature method with speeded up robust features for feature extraction and quadratic support vector machine for classification. The proposed method shows the accuracy of 85.7%, sensitivity of 100%, specificity of 60% and training time of 0.8507 s in classifying the lesion. The result and analysis of experiments are done on the PH2 dataset of skin cancer. Our method improves performance accuracy with an increase of 3% than other state-of-the-art methods.