Browsing by Author "Yalniz, I. Z."

Now showing 1 - 4 of 4

Open Access
Automatic detection and segmentation of orchards using very high resolution imagery
(Institute of Electrical and Electronics Engineers, 2012-08) Aksoy, S.; Yalniz, I. Z.; Tasdemir, K.
Spectral information alone is often not sufficient to distinguish certain terrain classes such as permanent crops like orchards, vineyards, and olive groves from other types of vegetation. However, instances of these classes possess distinctive spatial structures that can be observable in detail in very high spatial resolution images. This paper proposes a novel unsupervised algorithm for the detection and segmentation of orchards. The detection step uses a texture model that is based on the idea that textures are made up of primitives (trees) appearing in a near-regular repetitive arrangement (planting patterns). The algorithm starts with the enhancement of potential tree locations by using multi-granularity isotropic filters. Then, the regularity of the planting patterns is quantified using projection profiles of the filter responses at multiple orientations. The result is a regularity score at each pixel for each granularity and orientation. Finally, the segmentation step iteratively merges neighboring pixels and regions belonging to similar planting patterns according to the similarities of their regularity scores and obtains the boundaries of individual orchards along with estimates of their granularities and orientations. Extensive experiments using Ikonos and QuickBird imagery as well as images taken from Google Earth show that the proposed algorithm provides good localization of the target objects even when no sharp boundaries exist in the image data. © 2012 IEEE.
Open Access
Integrated segmentation and recognition of connected Ottoman script
(S P I E - International Society for Optical Engineering, 2009-11) Yalniz, I. Z.; Altingovde, I. S.; Güdükbay, Uğur; Ulusoy, Özgür
We propose a novel context-sensitive segmentation and recognition method for connected letters in Ottoman script. This method first extracts a set of segments from a connected script and determines the candidate letters to which extracted segments are most similar. Next, a function is defined for scoring each different syntactically correct sequence of these candidate letters. To find the candidate letter sequence that maximizes the score function, a directed acyclic graph is constructed. The letters are finally recognized by computing the longest path in this graph. Experiments using a collection of printed Ottoman documents reveal that the proposed method provides >90% precision and recall figures in terms of character recognition. In a further set of experiments, we also demonstrate that the framework can be used as a building block for an information retrieval system for digital Ottoman archives. © 2009 Society of Photo-Optical Instrumentation Engineers.
Open Access
Ottoman archives explorer: a retrieval system for digital Ottoman archives
(Association for Computing Machinery, 2009-12) Yalniz, I. Z.; Altingovde, I. S.; Güdükbay, Uğur; Ulusoy, Özgür
This article presents Ottoman Archives Explorer, a Content-Based Retrieval (CBR) system based on character recognition for printed and handwritten historical documents. Several methods for character segmentation and recognition stages are investigated. In particular, sliding-window and histogram segmentation methods are coupled with recognition approaches using spatial features, neural networks, and a graph-based model. The prototype system provides CBR of document images using both example-based queries and a virtual keyboard to construct query words. © 2009 ACM.
Open Access
Unsupervised detection and localization of structural textures using projection profiles
(Elsevier BV, 2010) Yalniz, I. Z.; Aksoy, S.
The main goal of existing approaches for structural texture analysis has been the identification of repeating texture primitives and their placement patterns in images containing a single type of texture. We describe a novel unsupervised method for simultaneous detection and localization of multiple structural texture areas along with estimates of their orientations and scales in real images. First, multi-scale isotropic filters are used to enhance the potential texton locations. Then, regularity of the textons is quantified in terms of the periodicity of projection profiles of filter responses within sliding windows at multiple orientations. Next, a regularity index is computed for each pixel as the maximum regularity score together with its orientation and scale. Finally, thresholding of this regularity index produces accurate localization of structural textures in images containing different kinds of textures as well as non-textured areas. Experiments using three different data sets show the effectiveness of the proposed method in complex scenes. © 2010 Elsevier Ltd. All rights reserved.