Article

A new distance measure based on generalized Image Normalized Cross-Correlation for robust video tracking and image recognition

A. Nakhmani is with the Department of Electrical and Computer Engineering, Boston University, Boston, MA.
Pattern Recognition Letters 02/2013; 34(3):315-321. DOI: 10.1016/j.patrec.2012.10.025

ABSTRACT We propose two novel distance measures, normalized between 0 and 1, and based on normalized cross-correlation for image matching. These distance measures explicitly utilize the fact that for natural images there is a high correlation between spatially close pixels. Image matching is used in various computer vision tasks, and the requirements to the distance measure are application dependent. Image recognition applications require more shift and rotation robust measures. In contrast, registration and tracking applications require better localization and noise tolerance. In this paper, we explore different advantages of our distance measures, and compare them to other popular measures, including Normalized Cross-Correlation (NCC) and Image Euclidean Distance (IMED). We show which of the proposed measures is more appropriate for tracking, and which is appropriate for image recognition tasks.

Download full-text

Full-text

Available from: Arie Nakhmani, Aug 25, 2015
0 Followers
 · 
317 Views
  • [Show abstract] [Hide abstract]
    ABSTRACT: We present four novel point-to-set distances defined for fuzzy or gray-level image data, two based on integration over -cuts and two based on the fuzzy distance transform. We explore their theoretical properties. Inserting the proposed point-to-set distances in existing definitions of set-to-set distances, among which are the Hausdorff distance and the Sum of Minimal Distances, we define a number of distances between fuzzy sets. These set distances are directly applicable for comparing graylevel images or fuzzy segmented objects, but also for detecting patterns and matching parts of images. The distance measures integrate shape and intensity/membership of observed entities, providing a highly applicable tool for image processing and analysis. Performance evaluation of derived set distances in real image processing tasks is conducted and presented. It is shown that the considered distances have a number of appealing theoretical properties and exhibit very good performance in template matching and object classification, for fuzzy segmented images as well as when applied directly on gray-level intensity images. Examples include recognition of hand written digits and identification of virus particles. The proposed set distances perform excellently on the MNIST digit classification task, achieving the best reported error rate for classification using only rigid body transformations and a kNN classifier.
    IEEE Transactions on Image Processing 10/2013; 23(1). DOI:10.1109/TIP.2013.2286904 · 3.11 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: The spatial distribution and pattern of local contact stresses within the knee joint during activities of daily living have not been fully investigated. The objective of this study was to determine if common contact stress patterns exist on the tibial plateaus of human knees during simulated gait. To test this hypothesis, we developed a novel normalized cross-correlation (NCC) algorithm and applied it to the contact stresses on the tibial plateaus of 12 human cadaveric knees subjected to multi-directional loads mimicking gait. The contact stress profiles at different locations on the tibial plateaus were compared, where regions with similar contact stress patterns were identified across specimens. Three consistent regional patterns were found, among them two most prominent contact stress patterns were shared by 9-12 of all the knees and the third pattern was shared by 6-8 knees. The first pattern was located at the posterior aspect of the medial tibial plateau and had a single peak stress that occurred during the early stance phase. The second pattern was located at the central-posterior aspects of the lateral plateau and consisted of two peak stresses coincident with the timing of peak axial force at early and late stance. The third pattern was found on the anterior aspect of cartilage-to-cartilage contact region on the medial plateau consisted of double peak stresses. The differences in the location and profile of the contact stress patterns suggest that the medial and lateral menisci function to carry load at different points in the gait cycle: with the posterior aspect of the medial meniscus consistently distributing load only during the early phase of stance, and the posterior aspect of the lateral meniscus consistently distributing load during both the early and late phases of stance. This novel approach can help identify abnormalities in knee contact mechanics and provide a better understanding of the mechanical pathways leading to post-traumatic osteoarthritis.
    Journal of Biomechanics 12/2013; DOI:10.1016/j.jbiomech.2013.11.042 · 2.50 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: When evolving from a normal state to failure, mechanical systems undergo a gradual degradation process. Due to the nonlinearity of damage accumulation, degradation data always exhibit a distinctive trend and random fluctuations. It makes the prediction of remaining useful life (RUL) very difficult and inaccurate. The phase space trajectory reconstructed from the time series of degradation data is capable of reliably elucidating the nonlinear degradation behavior. In this paper, a novel method based on the similarity of the phase space trajectory is proposed for estimating the RUL of mechanical systems. First, the reference degradation trajectories are built with historical degradation data using the phase space reconstruction. Second, the similarities between the current degradation trajectory and the reference degradation trajectories are measured with a normalized cross correlation indicator, which is determined solely by the trajectory shape and is not interfered with the scaling and shifting of the trajectory. Trajectory shape and degradation stage matching algorithms are combined to find the optimal segments in the reference degradation trajectories compared with the current degradation trajectory. Finally, the RULs corresponding to the optimal matching segments are subjected to weighted averaging to obtain the RUL of the current degradation process. The proposed method is evaluated utilizing both simulated data in stochastic degradation processes and experimental data measured on an actual pump. The results show that the predicted RULs are very close to the actual RUL.
    Expert Systems with Applications 11/2014; 42(5). DOI:10.1016/j.eswa.2014.10.041 · 1.97 Impact Factor
Show more