Video Concept Detection Based on Multiple Features and Classifiers Fusion

The rapid growth of multimedia content necessitates powerful technologies to filter, classify, index and retrieve video documents more efficiently. However, the essential bottleneck of image and video analysis is the problem of semantic gap that low level features extracted by computers always fail to coincide with high-level concepts interpreted by humans. In this paper, we present a generic scheme for the detection video semantic concepts based on multiple visual features machine learning. Various global and local low-level visual features are systelrtically investigated, and kernelbased learning method equips the concept detection system to explore the potential of these features. Then we combine the different features and sub-systen on both classifier-level and kernel-level fusion that contribute to a more robust system Our proposed system is tested on the TRECVID dataset. The resulted Mean Average Precision （MAP） score is rmch better than the benchmark perforrmnce, which proves that our concepts detection engine develops a generic model and perforrrs well on both object and scene type concepts.

Knowledge discovery method for feature-decision level fusion of multiple classifiers

To improve the performance of the multiple classifier system, a new method of feature-decision level fusion is proposed based on knowledge discovery. In the new method, the base classifiers operate on different feature spaces and their types depend on different measures of between-class separability. The uncertainty measures corresponding to each output of each base classifier are induced from the established decision tables （DTs） in the form of mass function in the Dempster-Shafer theory （DST）. Furthermore, an effective fusion framework is built at the feature-decision level on the basis of a generalized rough set model and the DST. The experiment for the classification of hyperspectral remote sensing images shows that the performance of the classification can be improved by the proposed method compared with that of plurality voting （PV）.

Dynamic weighted voting for multiple classifier fusion:a generalized rough set method

To improve the performance of multiple classifier system, a knowledge discovery based dynamic weighted voting （KD-DWV） is proposed based on knowledge discovery. In the method, all base classifiers may be allowed to operate in different measurement/feature spaces to make the most of diverse classification information. The weights assigned to each output of a base classifier are estimated by the separability of training sample sets in relevant feature space. For this purpose, some decision tables （DTs） are established in terms of the diverse feature sets. And then the uncertainty measures of the separability are induced, in the form of mass functions in Dempster-Shafer theory （DST）, from each DTs based on generalized rough set model. From the mass functions, all the weights are calculated by a modified heuristic fusion function and assigned dynamically to each classifier varying with its output. The comparison experiment is performed on the hyperspectral remote sensing images. And the experimental results show that the performance of the classification can be improved by using the proposed method compared with the plurality voting （PV）.

CLASSIFIER FUSION BASED ON EVIDENCE THEORY AND ITS APPLICATION IN FACE RECOGNITION

A multiple classifier fusion approach based on evidence combination is proposed in this paper. The individual classifier is designed based on a refined Nearest Feature Line (NFL),which is called Center-based Nearest Neighbor (CNN). CNN retains the advantages of NFL while it has relatively low computational cost. Different member classifiers are trained based on different feature spaces respectively. Corresponding mass functions can be generated based on proposed mass function determination approach. The classification decision can be made based on the combined evidence and better classification performance can be expected. Experimental results on face recognition provided verify that the new approach is rational and effective.

Belief Combination of Classifiers for Incomplete Data

Data with missing values,or incomplete information,brings some challenges to the development of classification,as the incompleteness may significantly affect the performance of classifiers.In this paper,we handle missing values in both training and test sets with uncertainty and imprecision reasoning by proposing a new belief combination of classifier(BCC)method based on the evidence theory.The proposed BCC method aims to improve the classification performance of incomplete data by characterizing the uncertainty and imprecision brought by incompleteness.In BCC,different attributes are regarded as independent sources,and the collection of each attribute is considered as a subset.Then,multiple classifiers are trained with each subset independently and allow each observed attribute to provide a sub-classification result for the query pattern.Finally,these sub-classification results with different weights(discounting factors)are used to provide supplementary information to jointly determine the final classes of query patterns.The weights consist of two aspects:global and local.The global weight calculated by an optimization function is employed to represent the reliability of each classifier,and the local weight obtained by mining attribute distribution characteristics is used to quantify the importance of observed attributes to the pattern classification.Abundant comparative experiments including seven methods on twelve datasets are executed,demonstrating the out-performance of BCC over all baseline methods in terms of accuracy,precision,recall,F1 measure,with pertinent computational costs.

Dempster Shafer distance-based multi-classifier fusion method for pig cough recognition

High precision pig cough recognition and low computational cost is of great importance for the realization of early warning of pig respiratory diseases.Numerous researchers have improved the recognition rate of pig cough sounds to a certain extent from feature selection and feature fusion perspectives.However,there is still a margin for the improvement in the accuracy and complexity of existing methods.Meanwhile,it is challenging to further enhance the precision of a single classifier.Therefore,this study proposed a multi-classifier fusion strategy based on Dempster Shafer distance(DS-distance)algorithm to increase the classification accuracy.Considering the engineering implementation,the machine learning with low computational complexity for fusion was chosen.First,three metrics of accuracy and diversity between classifiers were defined,including overall accuracy(OA),double fault(DF),and overall accuracy and double fault(OADF),for selecting the base classifiers.Subsequently,a two-step base classifier selection approach based on these metrics was proposed to make an optimized selection of features and classifiers.Finally,the proposed DS-distance algorithm was used to fuse the selected base classifiers to create a classification.The sound data collected in the pig barn verified the proposed algorithm.The experimental results revealed that the overall recognition accuracy of the proposed method could reach 98.76%,which was better than the existing methods.This study has achieved a high recognition accuracy through ensembled machine learning with low computational complexity.The proposed method provided an efficient way for the quick establishment of high precision pig cough recognition model in practice.

Visual attention based model for target detection in large-field images

It is of great significance to rapidly detect targets in large-field remote sensing images,with limited computation resources.Employing relative achievements of visual attention in perception psychology,this paper proposes a hierarchical attention based model for target detection.Specifically,at the preattention stage,before getting salient regions,a fast computational approach is applied to build a saliency map.After that,the focus of attention（FOA） can be quickly obtained to indicate the salient objects.Then,at the attention stage,under the FOA guidance,the high-level visual features of the region of interest are extracted in parallel.Finally,at the post-attention stage,by integrating these parallel and independent visual attributes,a decision-template based classifier fusion strategy is proposed to discriminate the task-related targets from the other extracted salient objects.For comparison,experiments on ship detection are done for validating the effectiveness and feasibility of the proposed model.

Combination of classifiers with incomplete frames of discernment

The methods for combining multiple classifiers based on belief functions require to work with a common and complete(closed)Frame of Discernment(Fo D)on which the belief functions are defined before making their combination.This theoretical requirement is however difficult to satisfy in practice because some abnormal(or unknown)objects that do not belong to any predefined class of the Fo D can appear in real classification applications.The classifiers learnt using different attributes information can provide complementary knowledge which is very useful for making the classification but they are usually based on different Fo Ds.In order to clearly identify the specific class of the abnormal objects,we propose a new method for combination of classifiers working with incomplete frames of discernment,named CCIF for short.This is a progressive detection method that select and add the detected abnormal objects to the training data set.Because one pattern can be considered as an abnormal object by one classifier and be committed to a specific class by another one,a weighted evidence combination method is proposed to fuse the classification results of multiple classifiers.This new method offers the advantage to make a refined classification of abnormal objects,and to improve the classification accuracy thanks to the complementarity of the classifiers.Some experimental results are given to validate the effectiveness of the proposed method using real data sets.

Multi-angle Face Detection Based on DP-Adaboost

Although important progresses have been already made in face detection,many false faces can be found in detection results and false detection rate is influenced by some factors,such as rotation and tilt of human face,complicated background,illumination,scale,cloak and hairstyle.This paper proposes a new method called DP-Adaboost algorithm to detect multi-angle human face and improve the correct detection rate.An improved Adaboost algorithm with the fusion of frontal face classifier and a profile face classifier is used to detect the multi-angle face.An improved horizontal differential projection algorithm is put forward to remove those non-face images among the preliminary detection results from the improved Adaboost algorithm.Experiment results show that compared with the classical Adaboost algorithm with a frontal face classifier,the textual DP-Adaboost algorithm can reduce false rate significantly and improve hit rate in multi-angle face detection.