Skeleton-Based Recognition of Shapes in Images via Longest Path Matching

Bal Bozkurt, Gülce
Chambers, Erin Wolf
Gasparovic, Ellen
Hu, Ruizhen
Leonard, Kathryn
Shaker, Matineh
Wenk, Carola
We present a novel image recognition method based on the Blum medial axis that identifies shape information present in unsegmented input images. Inspired by prior work matching from a library using only the longest path in the medial axis, we extract medial axes from shapes with clean contours and seek to recognize these shapes within “no isy” images. Recognition consists of matching longest paths from the segmented images into complicated geometric graphs, which are computed via edge detection on the (unsegmented) input images to obtain Voronoi diagrams associated to the edges. We present two approaches: one based on map-matching techniques using the weak Fréchet distance, and one based on a multiscale curve metric after reducing the Voronoi graphs to their minimum spanning trees. This paper serves as a proof of concept for this approach, using images from three shape databases with known segmentability (whale flukes, strawberries, and dancers). Our preliminary results on these images show promise, with both approaches correctly identifying two out of three shapes.


Image resolution upscaling in the wavelet domain using directional cycle spinning
Temizel, Alptekin (2005-10-01)
Cycle spinning, a technique mainly used for wavelet denoising, has also been shown to be successful toward image resolution upscaling in the wavelet domain. We propose a directional variant of the cycle spinning methodology. We obtain estimates of local edge orientation from a wavelet decomposition of the available low-resolution image and use this information to influence the choice of cycle spinning parameters that are employed for resolution upscaling. Our experimental results show that the proposed meth...
Yasaroglu, Yagiz; Alatan, Abdullah Aydın (2014-10-30)
A novel watermarking method is presented in which the data embedded into a 3D model is extracted from an arbitrary 2D view by using a perspective projective invariant. The data is embedded into 3D positions of selected interest points on a 3D mesh. Determining the interest point modification vectors for ensuring watermark detection constitutes an important part of the proposed method. Different watermark embedding schemes based on optimization of the watermark function are implemented and evaluated. Another...
Broadband absorption enhancement in an uncooled microbolometer infrared detector
Kebapci, Basak; Dervisoglu, Ozgecan; Battal, Enes; OKYAY, Ali Kemal; Akın, Tayfun (2014-05-08)
This paper introduces a method for a broadband absorption enhancement in the LWIR range (8-12 µm), in single layer microbolometer pixels with 35 μm pitch. For the first time in the literature, this study introduces a very simple and low cost approach to enhance the absorption by embedding plasmonic structures at the same level as the already existing metallic layer of a microbolometer pixel. The metal layer comprises the electrode and the arm structures on the body. Even though the periodicity of the plasmo...
Specular motion and 3D shape estimation.
Dövencioğlu, Nahide Dicle; Barla, P; Doerschner, K (2017-06-01)
Dynamic visual information facilitates three-dimensional shape recognition. It is still unclear, however, whether the motion information generated by moving specularities across a surface is congruent to that available from optic flow produced by a matte-textured shape. Whereas the latter is directly linked to the first-order properties of the shape and its motion relative to the observer, the specular flow, the image flow generated by a specular object, is less sensitive to the object's motion and is tight...
KIRAGI, H; Ersak, Aydın (1994-04-14)
In this paper an object recognition and localization system based on ultrasonic range imaging to be used in optically opaque environments is introduced. The system is especially designed for robotics applications. The ultrasonic image is acquired by scanning ultrasonic transducers in two dimensions above the area where objects are located. The features that are used for recognition and localization processes are extracted from the outermost boundaries of the objects present in the input scene. Experimental ...
Citation Formats
G. Bal Bozkurt et al., Skeleton-Based Recognition of Shapes in Images via Longest Path Matching. 2015, p. 99.