Application Of Template Matching Research Articles

The application of template matching and novel unsupervised and supervised machine learning methodology to forensic hair analysis

ABSTRACT Hair pigmentation is a valuable feature in forensic hair analysis and hair comparisons. Microscopic pigmentation features from participants' hair were classified and documented. The discriminating power of template matching, unsupervised and supervised machine learning methods were compared to determine which method best distinguished between participants and accurately assigned hairto an individual, based on pigmentation features. Template matching analysis revealed a higher frequency of inter-person matches compared to intra-person matches, and the unsupervised model’s predicted labels did not align with true labels. These findings suggest the presence of an unknown dimensionality beyond the observable pigmentation features used in forensic hair classification, highlighting the crucial role of forensic expertise. In contrast, supervised machine learning demonstrated superior accuracy and greater interpretability due to its training on labelled data, making it more appropriate for forensic applications. This research underscores the continued importance of human expertise, particularly in the initial classification of training data for supervised machine learning. The potential applications of supervised machine learning in forensic science include training, competency testing and rapid intelligence as a novel and innovative tool while advancing the field towards more objective methodologies.

Pushing the limit of earthquake detection with distributed acoustic sensing and template matching: a case study at the Brady geothermal field

Template matching has been widely applied in the detection of earthquakes and other seismic events due to its power in detecting weak signals. Recent studies using synthetics have shown that application of template matching to large-N arrays can potentially detect earthquakes substantially below the noise level. Here we apply template matching to the distributed acoustic sensing (DAS) data recorded in the Brady Hot Springs geothermal field, Nevada. Using 5 catalogued events, we detect 116 events and find 68 of them well below the noise level. We confirm 112 events are true earthquakes by examining the patterns of their sensor-to-sensor cross-correlation sections. This demonstrates that the combination of DAS and template matching has capability to detect microseismicity below the noise level, which is unusual for conventional seismic arrays and methods. With the updated catalogue, we observe a surge of earthquakes during the shutdown of a geothermal power plant nearby. In addition, the rapid increases in the downhole pressure record coincide with intense swarms of earthquakes. These observations show a strong correlation between the seismicity frequencies and the downhole pressure changes. Finally, we investigate several factors that may affect the detection performance and compare different strategies for spatial down-sampling, in order to provide helpful insights for future large-N design and data processing.

Rotation invariants of vector fields from orthogonal moments

Vector field images are a type of new multidimensional data that appear in many engineering areas. Although the vector fields can be visualized as images, they differ from graylevel and color images in several aspects. To analyze them, special methods and algorithms must be originally developed or substantially adapted from the traditional image processing area. In this paper, we propose a method for the description and matching of vector field patterns under an unknown rotation of the field. Rotation of a vector field is so-called total rotation, where the action is applied not only on the spatial coordinates but also on the field values. Invariants of vector fields with respect to total rotation constructed from orthogonal Gaussian–Hermite moments and Zernike moments are introduced. Their numerical stability is shown to be better than that of the invariants published so far. We demonstrate their usefulness in a real world template matching application of rotated vector fields.

Application of Template Matching for Improving Classification of Urban Railroad Point Clouds.

This study develops an integrated data-driven and model-driven approach (template matching) that clusters the urban railroad point clouds into three classes of rail track, contact cable, and catenary cable. The employed dataset covers 630 m of the Dutch urban railroad corridors in which there are four rail tracks, two contact cables, and two catenary cables. The dataset includes only geometrical information (three dimensional (3D) coordinates of the points) with no intensity data and no RGB data. The obtained results indicate that all objects of interest are successfully classified at the object level with no false positives and no false negatives. The results also show that an average 97.3% precision and an average 97.7% accuracy at the point cloud level are achieved. The high precision and high accuracy of the rail track classification (both greater than 96%) at the point cloud level stems from the great impact of the employed template matching method on excluding the false positives. The cables also achieve quite high average precision (96.8%) and accuracy (98.4%) due to their high sampling and isolated position in the railroad corridor.

Accuracy and numerical stability of high-order polar harmonic transforms

Invariant moments and transforms are a special kind of statistical measure designed to remain constant after some transformations, such as object rotation, scaling, translation or image illumination changes, in order to improve the reliability of a pattern recognition system. Polar harmonic transforms (PHTs) are one of such image descriptors based on a set of orthogonal projection bases, which are better rated in comparison with other moments and transforms because they possess less redundant information. PHTs suffer from geometric error and numerical integration error. Geometric error is caused when a square image is mapped into a unit circular disk, which cannot exactly match the square domain. Numerical integration error is caused when the integration is approximated by zeroth-order summation. We propose a computational framework based on numerical integration approach that reduces the geometric error and the numerical integration error simultaneously. The effect of the improved accuracy in the PHTs is analysed for rotation invariance, scale invariance and image noise. Various numerical experiments demonstrate that the effect of these errors is much more prominent in small images and, therefore the proposed method is suitable for applications involving small images, such as in optical character recognition and template-matching applications.

FPGA Design for Statistics-Inspired Approximate Sum-of-Squared-Error Computation in Multimedia Applications

This brief introduces a low-cost hardware design for approximate squaring functions which preserve maximum information content of the signals in template-matching applications. Analysis of signal statistics for two example applications, i.e., motion estimation and disparity estimation, is presented. This information is then specifically incorporated in the hardware design process to develop approximate squarers which outperform existing designs in hardware resource savings and performance while processing real-world data. Specifically, the proposed architectures make distinction between low- and high-entropy portions of the input data to intelligently trade off bit precision with hardware complexity. Mathematical and experimental results show mean-relative-error figures to be as low as 1.2% and the performance to be as good as conventional full-precision processing scenarios. Implementation results for current-generation six-input look up table (LUT) and four-input LUT FPGAs have been discussed in relation to the proposed design flow.

Spiral Coating of the Endothelial Caveolar Membranes as Revealed by Electron Tomography and Template Matching

Caveolae are invaginations of the plasma membrane involved in multiple cellular processes, including transcytosis. In this paper we present an extensive 3-D electron tomographic study of the endothelial caveolar system in situ. Analysis of large cellular volumes of (high-pressure frozen, freeze-substituted and epon-embedded) human umbilical vein endothelial cells (HUVECs) provided a notable view on the architecture of the caveolar system that comprises--as confirmed by 3-D immunolabeling for caveolin of 'intact' cells--bona fide caveolae, free plasmalemmal vesicles, racemose invaginations and free multi-caveolar bodies. Application of template matching to tomograms allowed the 3-D localization of caveolar membrane coatings in a robust manner. In this way we observed that bona fide endothelial caveolae, cryofixed and embedded in their cellular context, show a spiral organization of the coating as shown in the past for chemically fixed and freeze-etched caveolae from fibroblasts. Meticulous 3-D analysis further revealed that the coatings are distributed in triads of spirals over the caveolar bulb and neck. Remarkably, this coating distribution is consistently present over the membranes of the other members of the caveolar system in HUVECs. The novel observations that we present clarify the ultrastructural complexity of the 'intact' caveolar system, setting a detailed morphological basis for its functional diversity.

Development of template matching hardware and its high‐speed processing strategy

AbstractThe template matching technique is now applied to tracking of outdoor object and position measurement, and there is increasing demand for application of template matching based on the normalized correlation coefficient, which is robust against the variation of illumination. In order to apply the technique based on the normalization technique to practical application problems, however, high‐speed calculation of the normalized correlation coefficient is required. This paper intends to improve the speed by implementing by hardware the calculation of the normalized correlation coefficient, which requires a large amount of computation. The proposed method decomposes the normalized correlation coefficient calculation into five operational components, each of which is implementated by hardware. The speed of processing is improved being combined with a parallel processing with a high degree of freedom based on the local property owned by the template matching process. For further improvement of the speed, the application of the pyramid structure is considered. Hardware that can handle the pyramid structure is realized. An evaluation was tried using an actual machine. The performance of the developed hardware is evaluated. The effectiveness of the approach is verified by applying the method to the traffic flow measurement system, as an example of an actual outdoor system. © 2001 Scripta Technica, Electron Comm Jpn Pt 3, 84(11): 1–10, 2001

Recognition of Fruits by Image Processing - Application of Template Matching-

The object of the present study is to develop the algorism capable recognizing some sorts of particular fruits and vegetables, for the robots being operated in farm working, by using the images obtained from the visual sensors mounted thereon. In the researches reported up to the present, recognitions of fruits, by utilizing the difference of the reflectance spectroscopic characteristics between the fruits and plants’ portions other than the fruits themselves in question, have been studied. In these cases, special types of filters and need, and moreover, the particular filters, best suited for each sort of the various harvests, are required. Therefore, in the present report, by means of the recognition algorithms using the image processing techniques not requiring special pre-processing procedures, recognition experiments of tomatoes were carried out.