International Journal of Pattern Recognition and Artificial Intelligence

The importance of finding correct correspondences between two images is the major aspect in problems such as appearance-based robot localization and content-based image retrieval. Local feature matching has become a commonly used method to compare images, despite being highly probable that at least some of the matchings/correspondences it detects are incorrect. In this paper, we describe a novel approach to local feature matching, named Feature Group Matching (FGM), to select stable features and obtain a more reliable similarity value between two images. The proposed technique is demonstrated to be translational, rotational and scaling invariant. Experimental evaluation was performed on large and heterogeneous datasets of images using SIFT and SURF, the actual state-of-the-art feature extractors. Results show that FGM avoids almost 95% of incorrect matchings, reduces the visual aliasing (number of images considered similar) and increases both robotic localization and image retrieval accuracy on the average of 13%.

This paper presents a method for adapting the cost function in the Monge–Kantorovich Problem (MKP) to a classification task. More specifically, we introduce a criterion that allows to learn a cost function which tends to produce large distance values for elements belonging to different classes and small distance values for elements belonging to the same class. Under some additional constraints (one of them being the well-known Monge condition), we show that the optimization of this criterion writes as a linear programming problem. Experimental results on synthetic data show that the output optimal cost function provides good retrieval performances in the presence of two types of perturbations commonly found in histograms. When compared to a set of various commonly used cost functions, our optimal cost function performs as good as the best cost function of the set, which shows that it can adapt well to the task. Promising results are also obtained on real data for two-class image retrieval based on grayscale intensity histograms.

In spite of its simplicity, naive Bayesian learning has been widely used in many data mining applications. However, the unrealistic assumption that all features are equally important negatively impacts the performance of naive Bayesian learning. In this paper, we propose a new method that uses a Kullback–Leibler measure to calculate the weights of the features analyzed in naive Bayesian learning. Its performance is compared to that of other state-of-the-art methods over a number of datasets.

Together with classification of facial expressions, the rating of their intensities is of major interest. Classical supervised learning techniques require labeling of the intensities, which is labor intensive and requires expert knowledge, but nevertheless is not guaranteed to be objective. We propose a new approach to learn an intensity rating function which does not require expert knowledge, because it simplifies the labeling task by avoiding the difficulty of selecting an absolute intensity value and to keep the labeling consistent for the whole dataset. It is based on a novel kind of ground truth which we call Comparative Labeling. It specifies sample pairs for which the first element is desired to have a lower intensity than the second. We introduce a learning scheme to find an optimal intensity function in respect of the Comparative Labeling and propose performance measures to assess the quality of the learned function. The technique is applied to rate the intensity of facial expressions of posed pain. The evaluation results show that the learned function is well suited for determining dynamic intensity variation over time. We also assess the suitability of the rating as an inter-individual intensity measure by comparing it to the intensity ratings given by human observers.

Texture is an important feature that aids in identifying objects of interest or region of interest irrespective of the source of the image. In this paper, a novel and simple isopattern-based texture feature is introduced. Spatial gray scale dependencies represented by bit plane is analyzed for specific patterns and are accumulated in bins. These are scaled by half-normal weighting function to provide isopattern texture feature. The ability of this texture feature in capturing textural variations of the images despite the presence of illumination, scale and rotation is demonstrated by conducting texture analysis on Brodatz, OuTex texture datasets and its classification accuracy on Kylberg dataset. The results of these two experimentation indicate that the proposed textural feature picks variation in texture significantly and has a better texture classification accuracy of 98.26% when compared with the state-of-the-art features like Gabor, GLCM and LBP.

This paper presents a robust approach for recognition of thermal face images based on decision level fusion of 34 different region classifiers. The region classifiers concentrate on local variations. They use singular value decomposition (SVD) for feature extraction. Fusion of decisions of the region classifier is done by using majority voting technique. The algorithm is tolerant against false exclusion of thermal information produced by the presence of inconsistent distribution of temperature statistics which generally make the identification process difficult. The algorithm is extensively evaluated on UGC-JU thermal face database, and Terravic facial infrared database and the recognition performance are found to be 95.83% and 100%, respectively. A comparative study has also been made with the existing works in the literature.

Facial expressions convey personal characteristics and subtle emotional states. This paper presents a new framework for modeling subtle facial motions of different people with different types of expressions from high-resolution facial expression tracking data to synthesize new stylized subtle facial expressions. A conceptual facial motion manifold is used for a unified representation of facial motion dynamics from three-dimensional (3D) high-resolution facial motions as well as from two-dimensional (2D) low-resolution facial motions. Variant subtle facial motions in different people with different expressions are modeled by nonlinear mappings from the embedded conceptual manifold to input facial motions using empirical kernel maps. We represent facial expressions by a factorized nonlinear generative model, which decomposes expression style factors and expression type factors from different people with multiple expressions. We also provide a mechanism to control the high-resolution facial motion model from low-resolution facial video sequence tracking and analysis. Using the decomposable generative model with a common motion manifold embedding, we can estimate parameters to control 3D high resolution facial expressions from 2D tracking results, which allows performance-driven control of high-resolution facial expressions.

The unique iris pattern of each human eye is complex, but easily be scanned or captured by a camera. However, the high cost infrared iris scanners used for acquisition causes inconvenience to users by distance related constraints. This restricts its widespread use in real-time applications such as airports and banks. The images captured by cameras under visible wavelength are obstructed by the presence of reflections and shadows which requires additional attention. The main objective of this paper is to propose a secure biometric iris authentication system by fusion of RGB channel information from the real-time data captured under visible wavelength and varying light conditions. The proposed system is adapted to a real-time noisy iris dataset. The effectiveness of this proposed system was tested on two different color iris datasets, namely, a public database UBIRISv1 and a newly created database SSNDS which contains images captured with any digital/mobile camera of minimum 5MP under unconstrained environments. This system supports the cross sensor acquisition and successful iris segmentation from these unconstrained inputs. The features from each channel are extracted using log Gabor filter and a matching is performed using hamming distance based on two thresholds (inter and intra class variations). The performance quality of the proposed biometric system leads to the feasibility of a new cost-effective approach for any real-time application, which requires authentication to ensure quality service, enhance security, eliminate fraud, and maximize effectiveness.

Changes in light intensity and angle present a major challenge to the creation of reliable face recognition systems. The existence of bright regions and dark regions has been shown to have a serious negative impact on the performance of face recognition systems. This paper proposes a solution to this problem based on self-quotient image (SQI) processing method. In this method, bright and dark areas are processed separately without changing the essential characteristics of the image of the face. The dark and light areas are processed separately by SQI. Experimental results indicate that this Single-Light-Region and Single-Dark-Region SQI method removes the adverse effect of multi-bright and multi-dark areas better than competing methods.

In a multi-agent system, an agent may utilize its idle time to assist other agents in the system. Intent recognition is proposed to accomplish this with minimal communication. An agent performing recognition observes the tasks other agents are performing and, unlike the much studied field of plan recognition, the overall intent of an agent is recognized instead of a specific plan. The observing agent may use capabilities that it has not observed. A conceptual framework is proposed for intent recognition systems. An implementation of the conceptual framework is tested and evaluated. We hypothesize that using intent recognition in a multi-agent system increases utility (where utility is domain specific) and decreases the amount of communication. We test our hypotheses using the domain of Cow Herding, where agents attempt to herd cow agents into team corrals. A set of metrics, including task time and number of communications, is used to compare the performance of plan recognition and intent recognition. In our results, we find that intent recognition agents communicate fewer times than plan recognition agents. In addition, unlike plan recognition, when agents use the novel approach of intent recognition, they select unobserved actions to perform. Intent recognition agents were also able to outperform plan recognition agents by consistently scoring more points in the Cow Herding domain. This research shows that under certain conditions, an intent recognition system is more efficient than a plan recognition system. The advantage of intent recognition over plan recognition becomes more apparent in complex domains.