Abdulkerim Çapar

Concurrent Segmentation and Recognition with Shape-Driven Fast Marching Methods

We present a variational framework that integrates the statistical boundary shape models into a Level Set system that is capable of both segmenting and recognizing objects. Since we aim to recognize objects, we trace the active contour and stop it near real object boundaries while inspecting the shape of the contour instead of enforcing the contour to get a priori shape. We get the location of character boundaries and character labels at the system output. We developed a promising local front stopping scheme based on both image and shape information for fast marching systems. A new object boundary shape signature model, based on directional Gauss gradient filter responses, is also proposed. The character recognition system that employs the new boundary shape descriptor outperforms the other systems, based on well-known boundary signatures such as centroid distance, curvature etc

Gradient-based shape descriptors

This paper presents two shape descriptors which could be applied to both binary and grayscale images. The proposed algorithm utilizes gradient based features which are extracted along the object boundaries. We use two-dimensional steerable G-Filters (IEEE Trans Pattern Anal Mach Intell 19(6):545–563, 1997) to obtain gradient information at different orientations and scales, and then aggregate the gradients into a shape signature. The signature derived from the rotated object is circularly shifted version of the signature derived from the original object. This property is called the circular-shifting rule (Affine-invariant gradient based shape descriptor. Lecture notes in computer science. International workshop on multimedia contents Representation, Classification and Security, pp 514–521, 2006). The shape descriptor is defined as the Fourier transform of the signature. We also provide a distance measure for the proposed descriptor by taking the circular-shifting rule into account. The performance of the proposed descriptor is evaluated over two databases; one containing digits taken from vehicle license plates and the other containing MPEG-7 Core Experiment and Kimia shape data set. The experiments show that the devised method outperforms other well-known Fourier-based shape descriptors such as centroid distance and boundary curvature.

Affine invariant gradient based shape descriptor

This paper presents an affine invariant shape descriptor which could be applied to both binary and gray-level images. The proposed algorithm uses gradient based features which are extracted along the object boundaries. We use two-dimensional steerable G-Filters [1] to obtain gradient information at different orientations. We aggregate the gradients into a shape signature. The signatures derived from rotated objects are shifted versions of the signatures derived from the original object. The shape descriptor is defined as the Fourier transform of the signature. We also provide a distance definition for the proposed descriptor taking shifted property of the signature into account. The performance of the proposed descriptor is evaluated over a database containing license plate characters. The experiments show that the devised method outperforms other well-known Fourier-based shape descriptors such as centroid distance and boundary curvature.

Affine Invariant Shape Descriptors

This paper presents affine-invariant shape descriptor which could be applied to both binary and gray-level images. The proposed algorithm utilizes gradient based features which are extracted along the object boundaries. We use two-dimensional steerable G-Filters ([1]) to obtain gradient information at different orientations and scales. We aggregate the gradients into a shape signature. The signature derived from the rotated object is circularly shifted version of the signature derived from the original object. This property is called the circular-shifting rule ([2]). The shape descriptor is defined as the Fourier transform of the signature. We also provide a distance definition for the proposed descriptor taking the circular-shifting rule into account. The performance of the proposed descriptor is evaluated over the databases containing digits taken from vehicle license plates. The experiments show that the devised method outperforms other well-known Fourier-based shape descriptors such as centroid distance and boundary curvature.

A Turkish Handprint Character Recognition System

This paper presents a study for recognizing isolated Turkish handwritten uppercase letters. In the study, first of all, a Turkish Handprint Character Database has been created from the students in Istanbul Technical University (ITU). There are about 20000 uppercase and 7000 digit samples in this database. Several feature extraction and classification techniques are realized and combined to find the best recognition system for Turkish characters. Features, obtained from Karhunen-Loeve Transform, Zernike Moments, Angular Radial Transform and Geometric Features, are classified with Artificial Neural Networks, K-Nearest Neighbor, Nearest Mean, Bayes, Parzen and Size Dependent Negative Log-Likelihood methods. Geometric moments, which are suitable for Turkish characters, are formed. KLT features are fused with other features since KLT gives the best recognition rate but has no information about the shape of the character where other methods have. The fused features of KLT and ART classified by SDNLL gives the best result for Turkish characters in the experiments.

DeepMQ: A Deep Learning Approach Based Myelin Quantification in Microscopic Fluorescence Images

Oligodendrocytes wrap around the axons and form the myelin. Myelin facilitates rapid neural signal transmission. Any damage to myelin disrupts neuronal communication leading to neurological diseases such as multiple sclerosis (MS). There is no cure for MS. This is, in part, due to lack of an efficient method for myelin quantification during drug screening. In this study, an image analysis based myelin sheath detection method, DeepMQ, is developed. The method consists of a feature extraction step followed by a deep learning based binary classification module. The images, which were acquired on a confocal microscope contain three channels and multiple z-sections. Each channel represents either oligodendroyctes, neurons, or nuclei. During feature extraction, 26-neighbours of each voxel is mapped onto a 2D feature image. This image is, then, fed to the deep learning classifier, in order to detect myelin. Results indicate that 93.38% accuracy is achieved in a set of fluorescence microscope images of mouse stem cell-derived oligodendroyctes and neurons. To the best of authors’ knowledge, this is the first study utilizing image analysis along with machine learning techniques to quantify myelination.

A deep learning based approach for classification of CerbB2 tumor cells in breast cancer

This study proposes a unique approach to classify CerbB2 tumor cell scores in breast cancer based on deep learning models. Another contribution of the study is the creation of a dataset from original breast cancer tissues. On the purpose of training, validating and testing with deep learning models cell fragments were generated from sample tissue images. CerbB2 tumor scores were generated for the cell fragments were classified with high performance by the aid of convolutional neural networks (CNN).

Segmentation of precursor lesions in cervical cancer using convolutional neural networks

Cervical carcinoma is one of the frequently seen cancers in the world and in our country, develops from precursor lesions. These precursor lesions are analyzed by pathologists so that the diagnosis of the disease can be made. In this study, a system that performs automatic detection of pre-cancerous lesions was performed using the convolutional neural networks (CNNs). In the training phase, lesion recognition performance of the proposed system has reached 92%. Thereafter, whole image was segmented by using 60 × 60 pixel tiles during the training phase. After all, the precursor lesions were segmented with 81.71% Dice coefficient.

A multi-level thresholding based segmentation method for microscopic fluorescence in situ hybridization (FISH) images

Fluorescence in situ hybridization (FISH) technique widely used in cancer diagnosis is based on displaying chromosomal regions as FISH signals by staining with specific dyes. In this study, a new multi-level thresholding based FISH signal segmentation method is proposed for images produced by FISH technique. Cell nuclei are segmented on images, that are grabbed from fluorescence microcopes at high resolution, with adaptive thresholding, distance transform and watershed methods. FISH signals falling in cell boundaries are detected by applying multi-level thresholding and morphological post processes thanks to proposed segmentation method. It is observed that the detection rate of the proposed method on 49 FISH images taken from real patients, are higher than other widely used techniques in the literature.

A boundary based feature extraction method for G-banded chromosome classification

The most widely used cytogenetic method is G-banded karyotyping. A new feature extraction method is proposed for G-banded chromosome recognition. Chromosome features are mostly extracted on chromosome skeleton. The main innovation of the proposed method is extracting features around chromosome boundary contours, not the skeleton. The circular boundary contour signatures are extracted and applied to Discrete Fourier Transformation to get contour descriptor vector. Some chromoseme geometric features are added to this descriptor vector to form the main feature vector. These feature vectors are applied to different classfiers as input and the performances are compared with skeleton based techniques. Experiments show that the proposed method outperforms skeleton based methods dramatically.