Bag Of Features Research Articles

Diagnosis of breast tissue in mammography images based local feature descriptors

Breast cancer is one of the leading causes of death by cancer among women. The high mortality rates and the occurrence of this cancer worldwide show the importance of the investigation and development of means for the detection and early diagnosis of this disease. Computer-Aided Detection and Diagnosis systems have been developed to improve diagnostic accuracy by radiologists. This work proposes a method for discriminating patterns of malignancy and benignity of masses in digitized mammography images through the analysis of local features. The method comparatively applies the Scale-Invariant Feature Transform (SIFT), Speed Up Robust Feature (SURF), Oriented Fast and Rotated BRIEF (ORB) and Local Binary Pattern (LBP) descriptors for local feature extraction. These features are represented by a Bag of Features (BoF) model, applied to provide new representations of the data and to reduce its dimensionality. Finally, the features are used as input for the Support Vector Machine (SVM), Adaptive Boosting (Adaboost) and Random Forests (RF) classifiers to differentiate malignant and benign masses. The method obtained significant results, reaching 100% sensitivity, 99.65% accuracy and 99.24% specificity for benign and malignant mass classification.

AuthCom: Authorship verification and compromised account detection in online social networks using AHP-TOPSIS embedded profiling based technique

In view of the rise in security and privacy concern in social networks, there has been an inadvertent increase in research related to framing of appropriate measures to detect the security breaches in social networks. Cyber criminals are misusing social networking platforms for inappropriate and illegitimate purposes such as posting or sending of illegitimate content which a genuine user will rarely do. Hence, whenever a sensitive and unusual text is posted by a user, there is a need to authenticate whether it is posted by the legitimate owner of the account or some imposter who might have compromised the legitimate profile. The process of authentication called authorship verification helps to handle the same. In this paper, authorship verification has been performed using different textual features such as n-grams, Bag of words (BOW), stylometric and folksonomy features to examine the authorship of tweets posted by the users on the microblogging platform Twitter. Appropriate classification and statistical analysis techniques have been applied to compute different performance parameters. From the experimental analysis, an important observation found is that though char n-grams have an upper hand to other features, still other applicable measures such as word n-grams, BOW, stylometric and folksonomy features cannot be overlooked as each user maintained consistency in different set of features. Accordingly, different feature selection techniques have been used to rank and select best feature for each user. From the comparative analysis of various similarity and statistical based feature selection techniques it is observed that AHP weighted TOPSIS method surpassed others in terms of different performance parameters. Further computation as per ranked features helped to improve the result by achieving an overall average F-score value of 93.82%.

Dental Biometrics for Human Identification using Bag of Features

Dental Biometrics for Human Identification using Bag of Features

Classification and recognition scheme for vegetable pests based on the BOF-SVM model

The current detection technology for vegetable pests mainly relies on artificial statistics, which exists many shortages such as requiring a large amount of labor, low efficiency, feedback delay and artificial faults. By rapid detection and image processing technology targeting at vegetable pests, not only can reduce manpower and pesticide use, but also provide decision support for precise spraying and improve the quality of vegetables. Practical research achievements are still relatively lacking on the rapid identification technology based on image processing technology in vegetable pests. Given the above background, this paper presents a classification and recognition scheme based on the bag-of-words model and support vector machine (BOF-SVM) on four important southern vegetable pests including Whiteflies, Phyllotreta Striolata, Plutella Xylostella and Thrips. This paper consists of four sub-algorithms. The first sub-algorithm is to compute the character description of pest images based on scale-invariant feature transformation. The second sub-algorithm is to compute the visual vocabulary based on bag of features. The third sub-algorithm is to compute the classifier of pests based on support vector machines. The last one is to classify the pest images using the classifier. In this study, C++ and Python language were used as implementation technologies with OpenCV and LibSVM function library based on BOF-SVM classification algorithm. Experiments showed that the average recognition accuracy was 91.56% for a single image category judgment with 80 images from the real environment, and the average time was 0.39 seconds. This algorithm has achieved the ideal operating speed and precision. It can provide decision support for UAV precise spraying, and also has good application prospect in agriculture. Keywords: agricultural image processing, vegetable pests, classification, recognition, bag of features, support vector machine DOI: 10.25165/j.ijabe.20181103.3477 Citation: Xiao D Q, Feng J Z, Lin T Y, Pang C H, Ye Y W. Classification and recognition scheme for vegetable pests based on the BOF-SVM model. Int J Agric & Biol Eng, 2018; 11(3): 190–196.

Emotion recognition from facial expressions using hybrid feature descriptors

Here, a hybrid feature descriptor-based method is proposed to recognise human emotions from their facial expressions. A combination of spatial bag of features (SBoFs) with spatial scale-invariant feature transform (SBoF-SSIFT), and SBoFs with spatial speeded up robust transform are utilised to improve the ability to recognise facial expressions. For classification of emotions, K-nearest neighbour and support vector machines (SVMs) with linear, polynomial, and radial basis function kernels are applied. SBoFs descriptor generates a fixed length feature vector for all sample images irrespective of their size. Spatial SIFT and SURF features are independent of scaling, rotation, translation, projective transforms, and partly to illumination changes. A modified form of bag of features (BoFs) is employed by involving feature's spatial information for facial emotion recognition. The proposed method differs from conventional methods that are used for simple object categorisation without using spatial information. Experiments have been performed on extended Cohn–Kanade (CK+) and Japanese female facial expression (JAFFE) data sets. SBoF-SSIFT with SVM resulted in a recognition accuracy of 98.5% on CK+ and 98.3% on JAFFE data set. Images are resized through selective pre-processing, thereby retaining only the information of interest and reducing computation time.

Identificación de los cambios en los ángulos posturales con las cargas y modos de uso de las maletas escolares

Purpose: To determine the changes in postural angles with different loads and usage of school bags and the possible association with pain in school children and adolescents between 6 and 16 years of age.Methodology: A cross-sectional study realized in a school in Cota municipality, to determine the school bag features and identification of postural angles changes with different loads and usage of the school bags and the possible association with back pain. A software was used to analyze the photographs and to determine possible alterations of postural angles with different loads of school bags.Results: 159 children were included in this study. The average age was 10.79 (SD +/- 2.6) years, 45.9% males and 54% females. The average weight, height and body mass index (BMI) were 41.2 kg; 146.7 cm and 18.5 kg / m2 respectively. The school bag average weight was 4.2 kg, equivalent to 12.8% of body weight. The average weight of the bag in females was 4.3 kg (11.5% of body weight) and 4.2 kg (11.7% of body weight). The variation of the angles using the bag carrying the load identified, show a statistically significant difference in craniovertebralangle in all groups evaluated (7-9 years p=0.002, 10-11 years p= 11 years p= <0.001). 27.6% of children reported pain with the bag use. There are a statistically significant association between the presence of incidental pain and the girls OR = 2.22 (95% CI 1.04-4.74), the perception of usual load in the school bag OR = 1.6 (95% CI 1.04-2.48), the bag average weight OR = 1.36 (95% CI 1.06-1.75).Conclusion: 51% of school bags exceed 10% of body weight in a range of 1.07-10.2 kg. In many cases, the empty school bag could become 5% of the body weight of the child. The school bag average weight was 13.3% in children who report pain. The back pain was the most common type. The school bag average weight and the perception of usual load in the schoolbag, are may be related to the incidental pain.

Prediction of radiographic abnormalities by the use of bag-of-features and convolutional neural networks

This study evaluated the feasibility of bag-of-features (BOF) and convolutional neural networks (CNN) for computer-aided detection in distinguishing normal from abnormal radiographic findings. Computed thoracic radiographs of dogs were collected. For the purposes of this study, radiographic findings were used to distinguish between normal and abnormal in the following areas: (1) normal cardiac silhouette vs. cardiomegaly, (2) normal lung vs. abnormal lung patterns, (3) normal mediastinal position vs. mediastinal shift, (4) normal pleural space vs. pleural effusion, and (5) normal pleural space vs. pneumothorax. Images for training and testing the models consisted of ventrodorsal and lateral projection images of the same scale. The number of images used for each finding are as follow: 3142 for cardiomegaly (1571 normal and 1571 abnormal from 1143 dogs), 2086 for lung pattern (1043 normal and 1043 abnormal from 1247 dogs), 892 for mediastinal shift (446 normal and 446 abnormal from 387 dogs), 940 for pleural effusion (470 normal and 470 abnormal from 284 dogs), and 78 for pneumothorax (39 normal and 39 abnormal from 61 dogs). All data samples were divided so that 60% would be used for training the algorithms and 40% for testing the two models. The performance of the classifiers was evaluated by calculating the accuracy, sensitivity and specificity.The accuracy of both models ranged from 79.6% to 96.9% in the testing set. CNN showed higher accuracy (CNN; 92.9–96.9% and BOF; 79.6–96.9%) and sensitivity (CNN; 92.1–100% and BOF; 74.1–94.8%) than BOF. In conclusion, both BOF and CNN have potential to be useful for improving work efficiency by double reading.

Assessment of Machine Learning Algorithms for Automatic Benthic Cover Monitoring and Mapping Using Towed Underwater Video Camera and High-Resolution Satellite Images

Benthic habitat monitoring is essential for many applications involving biodiversity, marine resource management, and the estimation of variations over temporal and spatial scales. Nevertheless, both automatic and semi-automatic analytical methods for deriving ecologically significant information from towed camera images are still limited. This study proposes a methodology that enables a high-resolution towed camera with a Global Navigation Satellite System (GNSS) to adaptively monitor and map benthic habitats. First, the towed camera finishes a pre-programmed initial survey to collect benthic habitat videos, which can then be converted to geo-located benthic habitat images. Second, an expert labels a number of benthic habitat images to class habitats manually. Third, attributes for categorizing these images are extracted automatically using the Bag of Features (BOF) algorithm. Fourth, benthic cover categories are detected automatically using Weighted Majority Voting (WMV) ensembles for Support Vector Machines (SVM), K-Nearest Neighbor (K-NN), and Bagging (BAG) classifiers. Fifth, WMV-trained ensembles can be used for categorizing more benthic cover images automatically. Finally, correctly categorized geo-located images can provide ground truth samples for benthic cover mapping using high-resolution satellite imagery. The proposed methodology was tested over Shiraho, Ishigaki Island, Japan, a heterogeneous coastal area. The WMV ensemble exhibited 89% overall accuracy for categorizing corals, sediments, seagrass, and algae species. Furthermore, the same WMV ensemble produced a benthic cover map using a Quickbird satellite image with 92.7% overall accuracy.

Grapevine buds detection and localization in 3D space based on Structure from Motion and 2D image classification

In viticulture, there are several applications where 3D bud detection and localization in vineyards is a necessary task susceptible to automation: measurement of sunlight exposure, autonomous pruning, bud counting, type-of-bud classification, bud geometric characterization, internode length, and bud development stage. This paper presents a workflow to achieve quality 3D localizations of grapevine buds based on well-known computer vision and machine learning algorithms when provided with images captured in natural field conditions (i.e., natural sunlight and the addition of no artificial elements), during the winter season and using a mobile phone RGB camera. Our pipeline combines the Oriented FAST and Rotated BRIEF (ORB) for keypoint detection, a Fast Local Descriptor for Dense Matching (DAISY) for describing the keypoint, and the Fast Approximate Nearest Neighbor (FLANN) technique for matching keypoints, with the Structure from Motion multi-view scheme for generating consistent 3D point clouds. Next, it uses a 2D scanning window classifier based on Bag of Features and Support Vectors Machine for classification of 3D points in the cloud. Finally, the Density-Based Spatial Clustering of Applications with Noise (DBSCAN) for 3D bud localization is applied. Our approach resulted in a maximum precision of 1.0 (i.e., no false detections), a maximum recall of 0.45 (i.e. 45% of the buds detected), and a localization error within the range of 259–554 pixels (corresponding to approximately 3 bud diameters, or 1.5 cm) when evaluated over the whole range of user-given parameters of workflow components.

Learning bag-of-embedded-words representations for textual information retrieval

Word embedding models are able to accurately model the semantic content of words. The process of extracting a set of word embedding vectors from a text document is similar to the feature extraction step of the Bag-of-Features (BoF) model, which is usually used in computer vision tasks. This gives rise to the proposed Bag-of-Embedded Words (BoEW) model that can efficiently represent text documents overcoming the limitations of previously predominantly used techniques, such as the textual Bag-of-Words model. The proposed method extends the regular BoF model by a) incorporating a weighting mask that allows for altering the importance of each learned codeword and b) by optimizing the model end-to-end (from the word embeddings to the weighting mask). Furthermore, the BoEW model also provides a fast way to fine-tune the learned representation towards the information need of the user using relevance feedback techniques. Finally, a novel spherical entropy objective function is proposed to optimize the learned representation for retrieval using the cosine similarity metric.

Bag of recurrence patterns representation for time-series classification

Time-Series Classification (TSC) has attracted a lot of attention in pattern recognition, because wide range of applications from different domains such as finance and health informatics deal with time-series signals. Bag of Features (BoF) model has achieved a great success in TSC task by summarizing signals according to the frequencies of "feature words" of a data-learned dictionary. This paper proposes embedding the Recurrence Plots (RP), a visualization technique for analysis of dynamic systems, in the BoF model for TSC. While the traditional BoF approach extracts features from 1D signal segments, this paper uses the RP to transform time-series into 2D texture images and then applies the BoF on them. Image representation of time-series enables us to explore different visual descriptors that are not available for 1D signals and to treats TSC task as a texture recognition problem. Experimental results on the UCI time-series classification archive demonstrates a significant accuracy boost by the proposed Bag of Recurrence patterns (BoR), compared not only to the existing BoF models, but also to the state-of-the art algorithms.

Locality constrained encoding of frequency and spatial information for image classification

The bag-of-feature (BoF) model provides a way to construct high-level representation for image classification. Although spatial pyramid matching (SPM) has been incorporated into many of its extensions, these models intrinsically lack the mechanism to utilize frequency domain information. In this paper, we propose the locality-constrained encoding of frequency and spatial information (LEFSI) algorithm, in which an image is decomposed into multiple frequency components and each component is further decomposed into subregions using SPM. The scale-invariant feature transform (SIFT) descriptors are first calculated in each subregion, and then converted into a global descriptor by using the codebook generated on a category-by-category basis and locality-constrained linear coding (LLC). The image feature is defined as the concatenation of global descriptors constructed in all subregions. We evaluated this algorithm against several state-of-the-art models on six benchmark datasets. Our results suggest that the proposed LEFSI algorithm can describe images more effectively and provide more accurate image classification.

Bag of Shape Features with a learned pooling function for shape recognition

Bag of Shape Features (BoSF), such as Bag of Contour Fragments (BoCF) and Bag of Skeleton-associated Contour Parts (BoSCP), derived from the well-known Bag of Features (BoF), is an effective framework for shape representation. The feature pooling in this framework is a critical step, while either max pooling or average pooling is not a learnable process. In this paper, we aim at learning a pooling function which is adaptive to the input shape features instead. Towards this end, we formulate our pooling function as a weighted sum of max pooling and average pooling, where the weight is expressed by an activation function of the input shape features. To automatically learn this weight, the output of the pooling function is fed into a SVM classifier and they are trained jointly to minimize a shape classification loss. Experimental results on several standard shape datasets demonstrate the effectiveness of the proposed learned pooling function, which can achieve considerable improvements compared with both BoCF and BoSCP.

A Hybrid Methodology for Automation the Diagnosis of Leukemia Based on Quantitative and Morphological Feature Analysis

Recent year’s witnessed a huge revolution for developing an automated diagnosis for different disease such as cancer using medical image processing. Many researches have been dedicated to achieve this goal. Analyzing medical microscopic histology images provide us with large information about the status of patient and the progress of diseases, help to determine if the tissue have any pathological changes. Automation of the diagnosis of these images will lead to better, faster and enhanced diagnosis for different hematological and histological tissue images such as cancer. This paper propose an automated methodology for analyzing cancer histology and hematology microscopic images to detect leukemia using image processing by combining two diagnosis procedures initial and advance; the initial diagnosis depend on the percentage of the white blood cells in microscopic images affected by leukemia as indicator for the existence of leukemia in the blood smear sample. Whereas, the advance diagnosis classifying the leukemia according into different types using feature bag classifier. The experimental results showed that the proposed methodology initial diagnosis is able to detect leukemia images and differentiate it from samples that do not have leukemia. While, advance diagnosis it is able to detect and classify most leukemia types and differentiate between acute and chronic, but in some cases in the chronic leukemia where the percent of blast cells and shape are similar; it gave a diagnosis of the type of leukemia to the most similar type.

Full-Reference Objective Quality Assessment of Tone-Mapped Images

In this paper we present a novel method for full-reference image quality assessment (IQA) of tone-mapped images displayed on standard low dynamic range (LDR) displays. Due to the dynamic range compression caused by the tone-mapping process a mixture of several artifacts and distortions may be produced in the tone-mapped images. This makes the quality assessment of the tone-mapped images very challenging. Due to the diversity of such artifacts and distortions we propose a “bag of features” (BOF) approach to tackle this problem. Specifically in the proposed method a number of different perceptually relevant quality-related features are first extracted from a given tone-mapped image and its reference HDR image. These features are designed such that they capture different aspects and attributes of the tone-mapped image such as its structural fidelity naturalness and overall brightness. A support vector regressor is then trained based on the extracted features and it is used for measuring the visual quality of a tone-mapped image. Our experimental results indicate that the proposed method achieves high accuracy as compared to several existing methods.

Action recognition in depth videos using hierarchical gaussian descriptor

In this paper, we propose a new approach based on distribution descriptors for action recognition in depth videos. Our local features are computed from binary patterns which incorporate the shape and motion cues for effective action recognition. Given pixel-level features, our approach estimates video local statistics in a hierarchical manner, where the distribution of pixel-level features and that of frame-level descriptors are modeled using single Gaussians. In this way, our approach constructs video descriptors directly from low-level features without resorting to codebook learning required by Bag-of-features (BoF) based approaches. In order to capture the spatial geometry and temporal order of a video, we use a spatio-temporal pyramid representation for each video. Our approach is validated on six benchmark datasets, i.e. MSRAction3D, MSRGesture3D, DHA, SKIG, UTD-MHAD and CAD-120. The experimental results show that our approach gives good performance on all the datasets. In particular, it achieves state-of-the-art accuracies on DHA, SKIG and UTD-MHAD datasets.

RGBD co-saliency detection via multiple kernel boosting and fusion

RGBD co-saliency detection, which aims at extracting common salient objects from a group of RGBD images with the additional depth information, has become an emerging branch of saliency detection. In this regard, this paper proposes a novel framework via multiple kernel boosting (MKB) and co-saliency quality based fusion. First, on the basis of pre-segmented regions at multiple scales, the regional clustering by feature bagging is exploited to generate the base co-saliency maps. Then the clustering-based samples selection is performed to select the most similar regions with high saliency from different images in the image set. The selected samples are utilized to learn a MKB-based regressor, which is applied to all regions at multiple scales to generate the MKB-based co-saliency maps. Finally, to make full use of both MKB and clustering-based co-saliency maps, a co-saliency quality criterion is proposed for adaptive fusion to generate the final co-saliency maps. Experimental results on a public RGBD co-saliency detection dataset demonstrate that the proposed co-saliency model outperforms the state-of-the-art co-saliency models.

Defect Classification of Electronic Board Using Dense SIFT and CNN

This paper proposes a new defect classification method of electronic board using Dense SIFT and CNN which can represent the effective features to the gray scale image. Proposed method does not use any reference image and effective keypoints are detected using Dense SIFT on the defect candidate region. Removing the feature points except defect region and Bag of Features are used to represent the histogram features. Dense SIFT and SVM are used to judge defect or not. CNN is further introduced to classify true or pseudo defect. Classification accuracy was evaluated and effectiveness of the proposed method is shown.

Reassessing Hierarchical Representation for Action Recognition in Still Images

Typical still action recognition methods rely on human body part detection and object detection. However, current human body part detectors and object detectors are far from perfect, leading to a negative impact on subsequent spatial relation learning of human-object interactions (HOIs). Bag-of-features (BoF)-based methods go beyond such modeling paradigms, but they do not achieve the state-of-the-art accuracies. In this paper, we propose two still action recognition methods that model HOI layouts by image hierarchical representation, rather than explicitly constructing HOI relations. The first method encodes a dense set of SIFT features using Fisher vectors, where an image is divided into increasingly fine regions with the spatial pyramid. The second method takes recent pretrained deep networks as feature execrators, where an image is divided into overlapped regions. The improvement effect of the hierarchical representation is proven by extensive comparison experiments. Our methods are very simple and easy-to-use, which remarkably outperform those BoF-based methods and complicated human-centric methods. To the best of our knowledge, our methods achieve the highest accuracies to date on the Sports, PPMI, and extended PPMI data sets.

PCA REDUCED FOREST FOR LEARNING TO RANK SPOKEN TRANSCRIPTIONS

This paper discusses the problem of learning to rank specially for spoken transcriptions. The state-of-art approach for text/web documents is to apply machine learning techniques to learn a ranking model from labeled query-documents pairs with their features. One of the best state-of-art learning algorithms is the Random Forest, however it does not perform very well when features are dependent or are monotonic transformation of other features as this makes the trees of the forest less independent. We propose to use Principal Component Analysis (PCA) to bags of features, in order to reduce them to simplify the model and have a surrogate score for each field's features producing more independent set of features for the Random Forest. Using this technique for a transcriptions dataset, 4.32% improvement in terms of Expected Reciprocal Rank (ERR@10) and 0.4% improvement in terms of Normalized Discounted Cumulative Gain (NDCG@10) for training data are achieved with very comparable results for the testing data. We emphasized the effectiveness of the technique by applying it to the larger and benchmarked web documents dataset; Microsoft LETOR. An improvement of 7.99% and 1.29% for test data are achieved for the two used metrics respectively.