Chuancai Liu

Statistical thresholding method for infrared images

Conventional statistical thresholding methods use class variance sum as criterions for threshold selection. These approaches neglect specific characteristic of practical images and fail to obtain satisfactory results when segmenting some images with similar statistical distributions in the object and background. To eliminate the limitation, a novel statistical criterion is defined by utilizing standard deviations of two thresholded classes, and the optimal threshold is determined by optimizing the criterion. The proposed method was compared with several classic thresholding counterparts on a variety of infrared images as well as general real-world ones, and the experimental results demonstrate its superiority.

Unsupervised range-constrained thresholding

Three range-constrained thresholding methods are proposed in the light of human visual perception. The new methods first implement gray level range-estimation, using image statistical characteristics in the light of human visual perception. An image transformation is followed by virtue of estimated ranges. Criteria of conventional thresholding approaches are then applied to the transformed image for threshold selection. The key issue in the process lies in image transformation which is based on unsupervised estimation for gray level ranges of object and background. The transformation process takes advantage of properties of human visual perception and simplifies an original image, which is helpful for image thresholding. Three new methods were compared with their counterparts on a variety of images including nondestructive testing ones, and the experimental results show its effectiveness.

Minimum Standard Deviation Difference-Based Thresholding

Classic statistical thresholding methods do not consider special characteristic of practical images and fail to obtain satisfying segmentation results for some images with similar statistical distributions in the object and background. In this paper, a novel statistical thresholding method based on standard deviation difference is presented to solve this problem. The proposed method defines standard deviation difference as criterion for threshold selection, and determines the optimal threshold by minimizing it. The new method was compared with three conventional thresholding methods on a variety of infrared images and general real world images, and experimental results show its effectiveness.

Iterative optimization for frame-by-frame object pose tracking ☆

Abstract Joint object tracking and pose estimation is an important issue in Augmented Reality (AR), interactive systems, and robotic systems. Many studies are based on object detection methods that only focus on the reliability of the features. Other methods combine object detection with frame-by-frame tracking using the temporal redundancy in the video. However, in some mixed methods, the interval between consecutive detection frames is usually too short to take the full advantage of the frame-by-frame tracking, or there is no appropriate switching mechanism between detection and tracking. In this paper, an iterative optimization tracking method is proposed to alleviate the deviations of the tracking points and prolong the interval, and thus speed up the pose estimation process. Moreover, an adaptive detection interval algorithm is developed, which can make the switch between detection and frame-by-frame tracking automatically according to the quality of frames so as to improve the accuracy in a tough tracking environment. Experimental results on the benchmark dataset manifest that the proposed algorithms, as an independent part, can be combined with some inter-frame tracking methods for optimization.

A new image thresholding method based on isoperimetric ratios

This paper presents a new thresholding method based on isoperimetric theory for image segmentation. The proposed method uses isoperimetric ratio as a criterion in selecting optimal segmenting threshold, that is, segmenting image with the gray value of minimal isoperimetric ratio as optimal threshold. Experiments show that this method outperforms normalized cut thresholding method in terms of segmentation quality, speed and noise immunity.

A double circle structure descriptor and Hough voting matching for real-time object detection

In this paper, we propose real-time and reliable approaches for pose tracking of a rigid object by feature detection and image matching. We first present a new fast binary descriptor with a double circle structure of overlapping regions, namely double circle structure descriptor (DCSD). DCSD is rotation invariant and robust against blur, illumination changes, Joint Photographic Experts Group (JPEG) compression and orientation changes. Experimental results show that with fewer feature bits, DCSD is still discriminative and faster than the state-of-the-art features in many general situations. We then propose a new matching measure named Hough Voting Matching (HVM), which is based on clustering and Hough voting schemes. HVM can efficiently discriminate between correct and incorrect keypoint correspondences, and can be combined with some descriptors to improve the matching accuracy as an independent part. Experiments are also presented to illustrate that HVM can refine the matching results of DCSD if we embed HVM into a DCSD algorithm.

Improved frame-by-frame object pose tracking in complex environments

Pose tracking is an important task in Augmented Reality (AR), interactive systems, and robotic systems. The frame-by-frame pose tracking that is effective in many cases still faces challenges in complex environments such as occlusions, illumination changes and flipping. In this paper, based on the optimization model offered by Ye et al. J Vis Commun Image Represent 44:72–81 (2017), three improvements are further proposed. First, a feature adjustment strategy based on a group of neighbors is offered to alleviate a sharp reduction of features. Then, when the features are no longer well representing the scene of interest, a score model based on a weighted histogram for result evaluations is presented to realize an adaptive interval. Besides, a forward-backward algorithm is provided to improve the accuracy by replacing the detection method with the tracking method. Experimental results manifest the effectiveness of the proposed algorithms.

Histogram-based embedding for learning on statistical manifolds

AbstractA novel binning and learning framework is presented for analyzing and applying large data sets that have no explicit knowledge of distribution parameterizations, and can only be assumed generated by the underlying probability density functions (PDFs) lying on a nonparametric statistical manifold. For models’ discretization, the uniform sampling-based data space partition is used to bin flat-distributed data sets, while the quantile-based binning is adopted for complex distributed data sets to reduce the number of under-smoothed bins in histograms on average. The compactified histogram embedding is designed so that the Fisher–Riemannian structured multinomial manifold is compatible to the intrinsic geometry of nonparametric statistical manifold, providing a computationally efficient model space for information distance calculation between binned distributions. In particular, without considering histogramming in optimal bin number, we utilize multiple random partitions on data space to embed the associated data sets onto a product multinomial manifold to integrate the complementary bin information with an information metric designed by factor geodesic distances, further alleviating the effect of over-smoothing problem. Using the equipped metric on the embedded submanifold, we improve classical manifold learning and dimension estimation algorithms in metric-adaptive versions to facilitate lower-dimensional Euclidean embedding. The effectiveness of our method is verified by visualization of data sets drawn from known manifolds, visualization and recognition on a subset of ALOI object database, and Gabor feature-based face recognition on the FERET database.

Generalizing intersection kernel support vector machines for color texture based recognition

Abstract This paper presents a novel recognition approach in which the component-adaptive color co-occurrence matrices (CACCMs) are designed to characterize color and texture cues in the images, while histogram intersection kernel support vector machines (HIKSVMs) are generalized to the version compatible to color co-occurrence matrix (CCM), called CCM intersection kernel support vector machines (CIKSVMs). An ensemble learning framework is proposed for synchronously training the optimal marginal CIKSVMs and corresponding CACCMs’ extractors. This learning architecture is applicable to an arbitrary color space employed for image coding, while we pay utmost attention to a perceptual uniform color space for the prominent potential in image proprieties’ display. For the formulation of recognition algorithm, the set of multi-channel CACCMs (CAMCMs) of per sample is utilized to get a balance between discriminative power and computational efficiency, while multiple marginal CIKSVMs are combined by weighted majority voting. The effectiveness of our approach is validated by promising results obtained from four experimental datasets.