Multiclass Object Classification Research Articles

ID-NET:INCEPTION DECONVOLUTIONAL NEURAL NETWORK FOR MULTI-CLASS CLASSIFICATION IN RETINAL FUNDUS IMAGE

Micro-aneurysms, hemorrhages, and hard exudates are the early lesions of diabetic retinopathy, with micro-aneurysms and small hemorrhages appearing as small and multi-size objects in the retinal fundus image. Though it is essentially a promising technique for object classification in images, the deep learning method needs specific investigation for dealing with retinal fundus images. To this, we propose an inception deconvolutional network, dubbed ID-Net, for multi-class object classification. The ID-Net consists of two inception modules, one with a series of convolutional filters used for extracting multi-scale features and the other with a stack of deconvolutional filters for recovering the features and thus increasing the feature representation capability. As a result, the proposed network succeeds in overcoming the under-fitting and representational bottlenecks, two challenges caused by imbalanced data of retinal fundus images. Experimental results show that the ID-Net can accurately classify early lesions, and achieves higher sensitivity and accuracy than the state-of-the-art evaluation on the public database.

Integration of complex wavelet transform and Zernike moment for multi‐class classification

Multiclass object classification is a crucial problem in computer vision research and have different emerging applications such as video surveillance. The task of multiclass object classification has more challenges because of highly variable nature and real time processing requirement of data. For tackling the multiclass object classification task, several existing methods adopt one feature or combination of features to classify objects. In this work, we propose a new combination of features-based algorithm for object classification. In the combination, the two features: (1) Daubechies complex wavelet transform (DCxWT) and (2) Zernike moments (ZM) have been used. The shift-invariance and symmetry properties of DCxWT facilitate the object classification in the wavelet domain. Specifically, the shift-invariance property of DCxWT is effective for translated object representation whereas the symmetry property yields perfect reconstruction for retaining object boundaries (i.e., edges). Moreover, translation and rotation-invariance properties of ZM are especially beneficial for the representation of varying pose and orientation of the objects. For these reasons, the composite of the two features brings about significant synthesized benefits over each single feature and the other widely used features. The multi-class support vector machine classifier is used for classifying different objects. The proposed method has been tested on standard datasets as well as our own dataset prepared by authors of this paper. Experimental results demonstrated the significant outperformance of the proposed method through quantitative evaluations and also suggest that the proposed hybridization of features is preferable for the classification problem.

On fusion schemes for multiclass object classification with reject in a given ensemble of sources

Given ensemble of the different modality sources, we investigated two data fusion schemes for constructing metric-based multiclass classifiers with a reject option. The decisions made for the composite objects given by the collections of the source objects of the same class with one object per each source. The fusion schemes are based on using different compositions. The first scheme uses the composition of the decisions on the objects of the individual sources and this scheme is known as the fusion by the weighted majority vote or WMV scheme. The second scheme uses the composition in a form of the general dissimilarity measure between any pair of the composite objects. This scheme is original and it is called by GDM scheme. We constructed multiclass classifiers as the appropriate collections of NN or SVM elementary “class-vs-all” classifiers. We performed classification experiments in the ensemble of decorrelated components of face RGB images. We shown that GDM scheme gives fewer error rates with respect to WMV scheme.

Multiclass object classification for real-time video surveillance systems

Object classification in video is an important factor for improving the reliability of various automatic applications in video surveillance systems, as well as a fundamental feature for advanced applications, such as scene understanding. Despite extensive research, existing methods exhibit relatively moderate classification accuracy when tested on a large variety of real-world scenarios, or do not obey the real-time constraints of video surveillance systems. Moreover, their performance is further degraded in multi-class classification problems. We explore multi-class object classification for real-time video surveillance systems and propose an approach for classifying objects in both low and high resolution images (human height varies from a few to tens of pixels) in varied real-world scenarios. Firstly, we present several features that jointly leverage the distinction between various classes. Secondly, we provide a feature-selection procedure based on entropy gain, which screens out superfluous features. Experiments, using various classification techniques, were performed on a large and varied database consisting of ∼29,000 object instances extracted from 140 different real-world indoor and outdoor, near-field and far-field scenes having various camera viewpoints, which capture a large variety of object appearances under real-world environmental conditions. The insight raised from the experiments is threefold: the efficiency of our feature set in discriminating between classes, the performance improvement when using the feature selection method, and the high classification accuracy obtained on our real-time system on both DSP (TMS320C6415-6E3, 600 MHz) and PC (Quad Core Intel® Xeon® E5310, 2 × 4 MB Cache, 1.60 GHz, 1066 MHz) platforms.

A linear structured approach and a refined fitness function in genetic programming for multi-class object classification

This paper describes an approach to the use of genetic programming (GP) to multi-class object recognition problems. Rather than using the standard tree structures to represent evolved classifier programs which only produce a single output value that must be further translated into a set of class labels, this approach uses a linear structure to represent evolved programs, which use multiple target registers each for a single class. The simple error rate fitness function is refined and a new fitness function is introduced to approximate the true feature space of an object recognition problem. This approach is examined and compared with the tree based GP on three data sets providing object recognition problems of increasing difficulty. The results show that this approach outperforms the standard tree based GP approach on all the tasks investigated here and that the programs evolved by this approach are easier to interpret. The investigation into the extra target registers and program length results in heuristic guidelines for initially setting system parameters.

Using Gaussian distribution to construct fitness functions in genetic programming for multiclass object classification

This paper describes a new approach to the use of Gaussian distribution in genetic programming (GP) for multiclass object classification problems. Instead of using predefined multiple thresholds to form different regions in the program output space for different classes, this approach uses probabilities of different classes, derived from Gaussian distributions, to construct the fitness function for classification. Two fitness measures, overlap area and weighted distribution distance, have been developed. Rather than using the best evolved program in a population, this approach uses multiple programs and a voting strategy to perform classification. The approach is examined on three multiclass object classification problems of increasing difficulty and compared with a basic GP approach. The results suggest that the new approach is more effective and more efficient than the basic GP approach. Although developed for object classification, this approach is expected to be able to be applied to other classification problems.