BOF Model Research Articles

Classification of Mammography Images using Bag of Features Model

The BoF model is a tool widely used in the fields of classification and image search, itspopularity is largely based on its simplicity. As part of our work we will be interested in an imageclassification mammography based on the BOF model with SVM algorithm application and the SIFTdescriptor. These algorithms show performance, the results have been compared with other algorithms. Inthe first phase, the SVM was used to compare the result with the MLP. In the same work, two SIFT andORB descriptors were used. To assess the effectiveness of the proposed work, the classification rate(Acouracy) and the precision rate were calculated. The best classification rates are obtained by applyingthe MLP algorithm and using the SIFT descriptor on the base images (Ultra Sounde Brest Images Dataset)with a rate of 86% a small increase compared to SVM with a rate equal to 84%. Accuracy rate with 80%for MLP and 74% for SVM respectively.

Large-Scale Image Retrieval of Tourist Attractions Based on Multiple Linear Regression Equations

This paper presents an in-depth study and analysis of large-scale tourist attraction image retrieval using multiple linear regression equation approaches. This feature extraction method often relies on the partitioning of the grid and is only effective when the overall similarity of different images is high. The BOF model is borrowed from the method for text retrieval, which generally extracts the local features of an image by the scale-invariant feature transform algorithm and clusters them using k -means to obtain a low-dimensional visual dictionary and characterizes the image features with a histogram vector based on the visual dictionary. However, when there are many kinds of images, the dimensionality of the visual dictionary will be large and it is not convenient to construct the BOF model. The last fully connected layer is taken as the image feature, and it is dimensionalized by the principal component analysis method, and then, the low-dimensional feature index structure is constructed using the locality-sensitive hashing- (LSH-) based approximate nearest neighbor algorithm. The accuracy of our graph retrieval has increased by 8%. The advantages of feature extraction by a convolutional neural network and the high efficiency of a hash index structure in retrieval are used to solve the shortcomings of traditional methods in terms of accuracy and other aspects in image retrieval. The results show that compared with the above two algorithms, for most of the attractions, the method has a relatively obvious advantage in the accuracy of retrieval, and when there are few similar images of a particular attraction in the attraction image library, the accuracy of the query results is not much different from the first two methods.

Temporal logistic neural Bag-of-Features for financial time series forecasting leveraging limit order book data

Time series forecasting is a crucial component of many important applications, ranging from forecasting the stock markets to energy load prediction. The high-dimensionality, velocity and variety of the data collected in many of these applications pose significant and unique challenges that must be carefully addressed for each of them. In this work, a novel Temporal Logistic Neural Bag-of-Features approach, that can be used to tackle these challenges, is proposed. The proposed method can be effectively combined with deep neural networks, leading to powerful deep learning models for time series analysis. However, combining existing BoF formulations with deep feature extractors pose significant challenges: the distribution of the input features is not stationary, tuning the hyper-parameters of the model can be especially difficult and the normalizations involved in the BoF model can cause significant instabilities during the training process. The proposed method is capable of overcoming these limitations by a employing a novel adaptive scaling mechanism and replacing the classical Gaussian-based density estimation involved in the regular BoF model with a logistic kernel. The effectiveness of the proposed approach is demonstrated using extensive experiments on a large-scale limit order book dataset that consists of more than 4 million limit orders.

REQUIREMENTS FOR MAGNESIUM-CARBON REFRACTORIES FOR BOF LINING AND FEATURES OF STEELMAKING TECHNOLOGIES IN ITS USE

The aim of the work is to develop a computer model of BOP-process that takes into account the interrelation of the parameters of steel smelting with the processes of refractories wear and is aimed at improving the conversion efficiency of metal. To achieve this goal, the authors has carried out a theoretical analysis of the patterns of assimilation of refractories by the slag phase formed during BOF smelting, used X-ray structural, spectral and microscopic studies of samples of magnesium-carbon refractories selected from different zones of the converter lining. By the method of regression analysis of the heats, mathematical models have been obtained that describe the features of the destruction of magnesium-carbon refractories. According to the research results, the main factors affecting the internal mass transfer in refractories under the conditions of its contact with the liquid slag melt have been defined. The requirements for the physical and chemical properties of magnesium-carbon refractories of the BOF working layer and for the production conditions have been substantiated. Scientific novelty of the work is the creation of a mathematical model for describing the process of internal mass transfer between the liquid slag phase of BOF and refractory, which takes place in its pores. The influence of the refractory porosity, pore diameter, concentration of the slag components, interfacial tension, wettability of the refractory with the liquid slag phase on the rate of internal mass transfer processes in BOF lining has been determined. The practical significance of the studies is a BOF model developed on a PC and adapted for PJSC “Dniprovsky Metallurgical Plant”, taking into account the interrelation of the parameters of metal conversion with the processes of magnesium-carbon refractories destruction, which allows to predict and optimize the process results.

Requirements for magnesium-carbon refractories for bof lining and features of steelmaking technologies in its use

The aim of the work is to develop a computer model of BOP-process that takes into account the interrelation of the parameters of steel smelting with the processes of refractories wear and is aimed at improving the conversion efficiency of metal. To achieve this goal, the authors has carried out a theoretical analysis of the patterns of assimilation of refractories by the slag phase formed during BOF smelting, used X-ray structural, spectral and microscopic studies of samples of magnesium-carbon refractories selected from different zones of the converter lining. By the method of regression analysis of the heats, mathematical models have been obtained that describe the features of the destruction of magnesium-carbon refractories. According to the research results, the main factors affecting the internal mass transfer in refractories under the conditions of its contact with the liquid slag melt have been defined. The requirements for the physical and chemical properties of magnesium-carbon refractories of the BOF working layer and for the production conditions have been substantiated. Scientific novelty of the work is the creation of a mathematical model for describing the process of internal mass transfer between the liquid slag phase of BOF and refractory, which takes place in its pores. The influence of the refractory porosity, pore diameter, concentration of the slag components, interfacial tension, wettability of the refractory with the liquid slag phase on the rate of internal mass transfer processes in BOF lining has been determined. The practical significance of the studies is a BOF model developed on a PC and adapted for PJSC “Dniprovsky Metallurgical Plant”, taking into account the interrelation of the parameters of metal conversion with the processes of magnesium-carbon refractories destruction, which allows to predict and optimize the process results.

Texture image classification based on BoF model with multi-feature fusion

Texture image classification based on BoF model with multi-feature fusion

A hierarchical representation for human action recognition in realistic scenes

BoF statistic-based local space-time features action representation is very popular for human action recognition due to its simplicity. However, the problem of large quantization error and weak semantic representation decrease traditional BoF model’s discriminant ability when applied to human action recognition in realistic scenes. To deal with the problems, we investigate the generalization ability of BoF framework for action representation as well as more effective feature encoding about high-level semantics. Towards this end, we present two-layer hierarchical codebook learning framework for human action classification in realistic scenes. In the first-layer action modelling, superpixel GMM model is developed to filter out noise features in STIP extraction resulted from cluttered background, and class-specific learning strategy is employed on the refined STIP feature space to construct compact and descriptive in-class action codebooks. In the second-layer of action representation, LDA-Km learning algorithm is proposed for feature dimensionality reduction and for acquiring more discriminative inter-class action codebook for classification. We take advantage of hierarchical framework’s representational power and the efficiency of BoF model to boost recognition performance in realistic scenes. In experiments, the performance of our proposed method is evaluated on four benchmark datasets: KTH, YouTube (UCF11), UCF Sports and Hollywood2. Experimental results show that the proposed approach achieves improved recognition accuracy than the baseline method. Comparisons with state-of-the-art works demonstrates the competitive ability both in recognition performance and time complexity.

Image Retrieval with Saliency Object Weighted and Bag of Visual Pair

In model of image-retrieval based on bag-of-features, a large number of information is lost during the process of quantizing the high-dimensional SIFT features as visual words. It leads to the teeming of errors matching feature points. To address this problem, this paper couples the color information into the BoF model as a complement to sift features, two different features are extracted to represent the same interest region and a visual pair is obtained by quantizing the descriptor pair with two independent codebooks. What’s more, we weight similarity degree of images by using the saliency area of image to decrease the semantic gap between low-level features and high-level expressions. We show that such an approach can significantly outperform matching results compared with traditional model of bag-of-features on Holiday dataset. Furthermore, our strategy of weighting based saliency further improved our performance in image-retrieval.

Local Random Sparse Coding for Human Action Recognition in Wireless Sensor Networks

Recognizing human action in wireless sensor networks (WSN) has raised a great interest owing to the requirements of real-world applications. Recently, the bag-of-features model (BOF) has proved effective in human action recognition. In this paper, we propose a novel method named local random sparse coding (LRSC) for human action recognition in WSN based on the BOF model. The contribution is twofold. First, we utilize random projection (RP) technique for each feature vector to alleviate the curse of dimensionality. Second, we consider the locality of codebook and correspondingly propose to reconstruct the features using similar codewords. Our method is verified on the KTH and UCF Sports databases, and the experimental results demonstrate that our method achieves better results than that of previous methods on human action recognition in WSN.

Spatial adjacent bag of features with multiple superpixels for object segmentation and classification

In the paper we present a new Spatial Adjacent Bag of Features (SABOF) model, in which the spatial information is effectively integrated into the traditional BOF model to enhance the scene and object recognition performance. The SABOF model chooses the frequency of each keyword and the largest frequency of its neighboring pairs to construct the feature histogram. Using the feature histogram whose dimension is only twice larger than that of the original BOF model, the SABOF model drastically enhances the discrimination performance. Combining the Superpixel Adjacent Histogram (SAH) Fulkerson et al., 2009 [12] with multiple segmentations Pantofaru et al., 2008 [33] and Russell et al., 2006 [36], the SABOF method effectively deals with the segmentation and classification of objects with different sizes. Changing the segmentation scale parameter to obtain multiple superpixel segmentations and correspondingly adjusting the neighbor parameters of the SAH method multiple classifiers are trained so that, the SABOF method can fuse multiple results of these classifiers to obtain better classification performance than any single classifier. The superpixel-based conditional random field (CRF) is used to further improve the classification performance. The experimental results of scene classification and of object recognition and localization on classical data sets demonstrate the performance of the proposed model and algorithm.

Spatial Pooling of Heterogeneous Features for Image Classification

In image classification tasks, one of the most successful algorithms is the bag-of-features (BoFs) model. Although the BoF model has many advantages, such as simplicity, generality, and scalability, it still suffers from several drawbacks, including the limited semantic description of local descriptors, lack of robust structures upon single visual words, and missing of efficient spatial weighting. To overcome these shortcomings, various techniques have been proposed, such as extracting multiple descriptors, spatial context modeling, and interest region detection. Though they have been proven to improve the BoF model to some extent, there still lacks a coherent scheme to integrate each individual module together. To address the problems above, we propose a novel framework with spatial pooling of complementary features. Our model expands the traditional BoF model on three aspects. First, we propose a new scheme for combining texture and edge-based local features together at the descriptor extraction level. Next, we build geometric visual phrases to model spatial context upon complementary features for midlevel image representation. Finally, based on a smoothed edgemap, a simple and effective spatial weighting scheme is performed to capture the image saliency. We test the proposed framework on several benchmark data sets for image classification. The extensive results show the superior performance of our algorithm over the state-of-the-art methods.