Robust Segmentation Algorithm Research Articles

Recognizing handwritten Arabic words using grapheme segmentation and recurrent neural networks

The Arabic alphabet is used in around 27 languages, including Arabic, Persian, Kurdish, Urdu, and Jawi. Many researchers have developed systems for recognizing cursive handwritten Arabic words, using both holistic and segmentation-based approaches. This paper introduces a system that achieves high accuracy using efficient segmentation, feature extraction, and recurrent neural network (RNN). We describe a robust rule-based segmentation algorithm that uses special feature points identified in the word skeleton to segment the cursive words into graphemes. We show that careful selection from a wide range of features extracted during and after the segmentation stage produces a feature set that significantly reduces the label error. We demonstrate that using same RNN recognition engine, the segmentation approach with efficient feature extraction gives better results than a holistic approach that extracts features from raw pixels. We evaluated this segmentation approach against an improved version of the holistic system MDLSTM that won the ICDAR 2009 Arabic handwritten word recognition competition. On the IfN/ENIT database of handwritten Arabic words, the segmentation approach reduces the average label error by 18.5 %, the sequence error by 22.3 %, and the execution time by 31 %, relative to MDLSTM. This approach also has the best published accuracies on two IfN/ENIT test sets.

Microscopy image segmentation tool: Robust image data analysis

We present a software package called Microscopy Image Segmentation Tool (MIST). MIST is designed for analysis of microscopy images which contain large collections of small regions of interest (ROIs). Originally developed for analysis of porous anodic alumina scanning electron images, MIST capabilities have been expanded to allow use in a large variety of problems including analysis of biological tissue, inorganic and organic film grain structure, as well as nano- and meso-scopic structures. MIST provides a robust segmentation algorithm for the ROIs, includes many useful analysis capabilities, and is highly flexible allowing incorporation of specialized user developed analysis. We describe the unique advantages MIST has over existing analysis software. In addition, we present a number of diverse applications to scanning electron microscopy, atomic force microscopy, magnetic force microscopy, scanning tunneling microscopy, and fluorescent confocal laser scanning microscopy.

Adaptive scale fuzzy local Gaussian mixture model for brain MR image segmentation

The Gaussian mixture model (GMM) has been widely used in brain magnetic resonance (MR) image segmentation. However, due to the MR bias field effect, the implied stochastic assumption that the intensities of each tissue type are sampled from an identical distribution may not be valid. In this paper, we propose a novel adaptive scale fuzzy local GMM (AS-FLGMM) algorithm for accurate and robust brain MR image segmentation. We assume that the local image data within the neighborhood of each pixel follow the GMM, in which the difference of variance among Gaussian components can be ignored. Based on this assumption, we develop a local scale estimation method to adaptively calculate the variance in each distribution. The segmentation is then performed under the fuzzy clustering framework and the objective is defined as the integration of the weighted GMM energy of each pixel. The AS-FLGMM algorithm has been compared to five state-of-the-art segmentation approaches in both synthetic and clinical MR images. Our results show that the proposed algorithm can produce more accurate segmentation results and its performance is more robust to initialization.

Brachial artery vasomotion and transducer pressure effect on measurements by active contour segmentation on ultrasound

To use feed-forward active contours (snakes) to track and measure brachial artery vasomotion on ultrasound images recorded in both transverse and longitudinal views; and to compare the algorithm's performance in each view. Longitudinal and transverse view ultrasound image sequences of 45 brachial arteries were segmented by feed-forward active contour (FFAC). The segmented regions were used to measure vasomotion artery diameter, cross-sectional area, and distention both as peak-to-peak diameter and as area. ECG waveforms were also simultaneously extracted frame-by-frame by thresholding a running finite-difference image between consecutive images. The arterial and ECG waveforms were compared as they traced each phase of the cardiac cycle. FFAC successfully segmented arteries in longitudinal and transverse views in all 45 cases. The automated analysis took significantly less time than manual tracing, but produced superior, well-behaved arterial waveforms. Automated arterial measurements also had lower interobserver variability as measured by correlation, difference in mean values, and coefficient of variation. Although FFAC successfully segmented both the longitudinal and transverse images, transverse measurements were less variable. The cross-sectional area computed from the longitudinal images was 27% lower than the area measured from transverse images, possibly due to the compression of the artery along the image depth by transducer pressure. FFAC is a robust and sensitive vasomotion segmentation algorithm in both transverse and longitudinal views. Transverse imaging may offer advantages over longitudinal imaging: transverse measurements are more consistent, possibly because the method is less sensitive to variations in transducer pressure during imaging.

Projected Image Restoration Method Based on Genetic Algorithm for Cone-Beam CT

For the problem of image quality degradation in cone-beam CT (CBCT) based on flat panel detector (FPD), a genetic algorithm based on pre-segmentation (PS-GA) is proposed for CBCT projected image restoration. According to the characteristic of that most of the area of the projected image is empty and without the tested object, a robust segmentation algorithm is used in this method to segment the smallest rectangle that contains the tested object, and the calculating range is limited to the smallest rectangle by the specially designed genetic algorithm, which significantly reduced the amount of calculated data. The experimental results show that the method raised the edge sharpness, contrast-to-noise ratio (CNR) and average gradient (AG) of the projected images and slice images, and there is no visible artifacts introduced.

Type-2 Fuzzy C-Means Algorithm with Spatial Informations and Membership Constrains for Liver Segmentation

The Fuzzy C-Means algorithm with spatial informations and membership constrains is a very effective and efficient image segmentation method. However£¬it is founded with Type-1 fuzzy sets, which can not handle the uncertainties existing in liver image well.The type-2 fuzzy sets have better performance on handling uncertainties than Type-1 fuzzy set. In this paper, a new robust Type-2 FCM image segmentation algorithm is proposed aiming to improve the segmentation precision and robustness of liver image by introducing the type-2 fuzzy set into FCM with spatial information and membership constrains. We extend the type-1 fuzzy set of membership to interval type-2 fuzzy set using two fuzzifiers and which create a footprint of uncertainty (FOU). The experimental results show that the target area of the liver in CT images can be segmented well by the proposed method.

Deformable templates guided discriminative models for robust 3D brain MRI segmentation.

Automatically segmenting anatomical structures from 3D brain MRI images is an important task in neuroimaging. One major challenge is to design and learn effective image models accounting for the large variability in anatomy and data acquisition protocols. A deformable template is a type of generative model that attempts to explicitly match an input image with a template (atlas), and thus, they are robust against global intensity changes. On the other hand, discriminative models combine local image features to capture complex image patterns. In this paper, we propose a robust brain image segmentation algorithm that fuses together deformable templates and informative features. It takes advantage of the adaptation capability of the generative model and the classification power of the discriminative models. The proposed algorithm achieves both robustness and efficiency, and can be used to segment brain MRI images with large anatomical variations. We perform an extensive experimental study on four datasets of T1-weighted brain MRI data from different sources (1,082 MRI scans in total) and observe consistent improvement over the state-of-the-art systems.

Video Object Segmentation and Tracking Framework With Improved Threshold Decision and Diffusion Distance

Video object segmentation and tracking are two essential building blocks of smart surveillance systems. However, there are several issues that need to be resolved. Threshold decision is a difficult problem for video object segmentation with a multibackground model. In addition, some conditions make robust video object tracking difficult. These conditions include nonrigid object motion, target appearance variations due to changes in illumination, and background clutter. In this paper, a video object segmentation and tracking framework is proposed for smart cameras in visual surveillance networks with two major contributions. First, we propose a robust threshold decision algorithm for video object segmentation with a multibackground model. Second, we propose a video object tracking framework based on a particle filter with the likelihood function composed of diffusion distance for measuring color histogram similarity and motion clue from video object segmentation. The proposed framework can track nonrigid moving objects under drastic changes in illumination and background clutter. Experimental results show that the presented algorithms perform well for several challenging sequences, and our proposed methods are effective for the aforementioned issues.

Automated three-dimensional choroidal vessel segmentation of 3D 1060 nm OCT retinal data

A fully automated, robust vessel segmentation algorithm has been developed for choroidal OCT, employing multiscale 3D edge filtering and projection of “probability cones” to determine the vessel “core”, even in the tomograms with low signal-to-noise ratio (SNR). Based on the ideal vessel response after registration and multiscale filtering, with computed depth related SNR, the vessel core estimate is dilated to quantify the full vessel diameter. As a consequence, various statistics can be computed using the 3D choroidal vessel information, such as ratios of inner (smaller) to outer (larger) choroidal vessels or the absolute/relative volume of choroid vessels. Choroidal vessel quantification can be displayed in various forms, focused and averaged within a special region of interest, or analyzed as the function of image depth. In this way, the proposed algorithm enables unique visualization of choroidal watershed zones, as well as the vessel size reduction when investigating the choroid from the sclera towards the retinal pigment epithelium (RPE). To the best of our knowledge, this is the first time that an automatic choroidal vessel segmentation algorithm is successfully applied to 1060 nm 3D OCT of healthy and diseased eyes.

Full Object Boundary Detection by Applying Scale Invariant Features in A Region Merging Segmentation Algorithm

Object detection is a fundamental task in computer vision and has many applications in image processing. This paper proposes a new approach for object detection by applying scale invariant feature transform (SIFT) in an automatic segmentation algorithm. SIFT is an invariant algorithm respect to scale, translation and rotation. The features are very distinct and provide stable keypoints that can be used for matching an object in different images. At first, an object is trained with different aspects for finding best keypoints. The object can be recognized in the other images by using achieved keypoints. Then, a robust segmentation algorithm is used to detect the object with full boundary based on SIFT keypoints. In segmentation algorithm, a merging role is defined to merge the regions in image with the assistance of keypoints. The results show that the proposed approach is reliable for object detection and can extract object boundary well.

Robust Automatic Continuous Speech Segmentation for Indian Languages to Improve Speech to Speech Translation

paper provides an analysis of phrase and word boundary detection in a background of noise, which occurs in the context of Automatic Recognition System (ASR) and Text- To-Speech (TTS) synthesis systems for Indian languages. ASR and TTS are the major components in Speech To Speech Translation (STST) system. Both are always need a speech signal to be segmented into some basic units like phrases, words, phonemes and syllables. Normal speech is a continuous sequence of sounds with no specific pause to indicate word boundaries. Hence to convert speech into corresponding text, it is necessary to identify the boundaries and phrases present in the continuous speech signal. In this work a robust algorithm for automatic continuous speech segmentation for Indian languages using short time energy and zero crossing rates has been proposed. This proposed method has been tested on various speakers' speech in four different Indian languages such as Tamil, Telugu, Hindi and Malayalam. The results shown to be computationally efficient for real time applications and it performs better than conventional methods for speech samples collected from noisy as well as noise free environment.

Illustration of the obstacles in computerized lung segmentation using examples

Automated lung volume segmentation is often a preprocessing step in quantitative lung computed tomography (CT) image analysis. The objective of this study is to identify the obstacles in computerized lung volume segmentation and illustrate those explicitly using real examples. Awareness of these "difficult" cases may be helpful for the development of a robust and consistent lung segmentation algorithm. We collected a large diverse dataset consisting of 2768 chest CT examinations acquired on 2292 subjects from various sources. These examinations cover a wide range of diseases, including lung cancer, chronic obstructive pulmonary disease, human immunodeficiency virus, pulmonary embolism, pneumonia, asthma, and interstitial lung disease (ILD). The CT acquisition protocols, including dose, scanners, and reconstruction kernels, vary significantly. After the application of a "neutral" thresholding-based approach to the collected CT examinations in a batch manner, the failed cases were subjectively identified and classified into different subgroups. Totally, 121 failed examinations are identified, corresponding to a failure ratio of 4.4%. These failed cases are summarized as 11 different subgroups, which is further classified into 3 broad categories: (1) failure caused by diseases, (2) failure caused by anatomy variability, and (3) failure caused by external factors. The failure percentages in these categories are 62.0%, 32.2%, and 5.8%, respectively. The presence of specific lung diseases (e.g., pulmonary nodules, ILD, and pneumonia) is the primary issue in computerized lung segmentation. The segmentation failures caused by external factors and anatomy variety are relatively low but unavoidable in practice. It is desirable to develop robust schemes to handle these issues in a single pass when a large number of CT examinations need to be analyzed.

Segmentation and Tracing of Single Neurons from 3D Confocal Microscope Images

In order to understand the brain, we need to first understand the morphology of neurons. In the neurobiology community, there have been recent pushes to analyze both neuron connectivity and the influence of structure on function. Currently, a technical road block that stands in the way of these studies is the inability to automatically trace neuronal structure from microscopy. On the image processing side, proposed tracing algorithms face difficulties in low contrast, indistinct boundaries, clutter, and complex branching structure. To tackle these difficulties, we develop Tree2Tree, a robust automatic neuron segmentation and morphology generation algorithm. Tree2Tree uses a local medial tree generation strategy in combination with a global tree linking to build a maximum likelihood global tree. Recasting the neuron tracing problem in a graph-theoretic context enables Tree2Tree to estimate bifurcations naturally, which is currently a challenge for current neuron tracing algorithms. Tests on cluttered confocal microscopy images of Drosophila neurons give results that correspond to ground truth within a margin of ±2.75% normalized mean absolute error.

The effect of smoking on early chorionic villous vascularisation

The aim of the study was to investigate whether first trimester chorionic villous vascularisation is different in women who smoked cigarettes before and during pregnancy in comparison with women who did not smoke.Placentas of smoking (>10 cigarettes/day, n = 13) and non-smoking women (n = 13), scheduled for a legal termination of a viable first trimester pregnancy for social indications, were retrieved. Placental tissues of 3–5 mm3 were whole mount CD31 immunofluorescence stained. Images of the CD31 immunofluorescence and contour of the villi were captured using an Optical Projection Tomography scanner. An immersive BARCO virtual reality system was used to create an enlarged interactive 3-dimensional hologram of the reconstructed images. Automatic volume measurements were performed using a flexible and robust segmentation algorithm that is based on a region-growing approach in combination with a neighbourhood variation threshold. The villous volume, vascular volume and vascular density were measured for the total chorionic villous tree as well as for its central and peripheral parts. No differences in maternal age and gestational age were found between non-smoking and smoking women. No differences were found in the total, central and peripheral villous tree volume and vascular volume. The central (13.4% vs. 9.5%, p=0.03) and peripheral (8.4% vs. 6.4%, p=0.02) villous tree vascular densities were increased in the smoking women as compared with the non-smoking women.In conclusion, chorionic villous vascularisation is already altered in first trimester of pregnancy in women who smoked cigarettes before and during pregnancy.

Unified Framework for Automated Iris Segmentation Using Distantly Acquired Face Images

Remote human identification using iris biometrics has high civilian and surveillance applications and its success requires the development of robust segmentation algorithm to automatically extract the iris region. This paper presents a new iris segmentation framework which can robustly segment the iris images acquired using near infrared or visible illumination. The proposed approach exploits multiple higher order local pixel dependencies to robustly classify the eye region pixels into iris or noniris regions. Face and eye detection modules have been incorporated in the unified framework to automatically provide the localized eye region from facial image for iris segmentation. We develop robust postprocessing operations algorithm to effectively mitigate the noisy pixels caused by the misclassification. Experimental results presented in this paper suggest significant improvement in the average segmentation errors over the previously proposed approaches, i.e., 47.5%, 34.1%, and 32.6% on UBIRIS.v2, FRGC, and CASIA.v4 at-a-distance databases, respectively. The usefulness of the proposed approach is also ascertained from recognition experiments on three different publicly available databases.

Color image segmentation using wavelet

In this paper, we discussed color image segmentation by extract the optimal features with which to discriminate between regions. Many real or texture images are made up of smooth regions and are best segmented using features in different areas. Schemas that select the optimal features for each pixel using wavelet analysis are proposed, leading to robust segmentation algorithm. Using two dimensions wavelet transforms to decompose the image into subbands channels and made up the of smooth image and convert the image into NTSC color space enables us to quantify the visual differences in the image, and then applies a clustering technique to partition the image into a set of “homogeneous” regions is also proposed.

Interactive Image Segmentation Based on Level Sets of Probabilities

In this paper, we present a robust and accurate algorithm for interactive image segmentation. The level set method is clearly advantageous for image objects with a complex topology and fragmented appearance. Our method integrates discriminative classification models and distance transforms with the level set method to avoid local minima and better snap to true object boundaries. The level set function approximates a transformed version of pixelwise posterior probabilities of being part of a target object. The evolution of its zero level set is driven by three force terms, region force, edge field force, and curvature force. These forces are based on a probabilistic classifier and an unsigned distance transform of salient edges. We further propose a technique that improves the performance of both the probabilistic classifier and the level set method over multiple passes. It makes the final object segmentation less sensitive to user interactions. Experiments and comparisons demonstrate the effectiveness of our method.

Comparison of parameter‐adapted segmentation methods for fluorescence micrographs

Interpreting images from fluorescence microscopy is often a time-consuming task with poor reproducibility. Various image processing routines that can help investigators evaluate the images are therefore useful. The critical aspect for a reliable automatic image analysis system is a robust segmentation algorithm that can perform accurate segmentation for different cell types. In this study, several image segmentation methods were therefore compared and evaluated in order to identify the most appropriate segmentation schemes that are usable with little new parameterization and robustly with different types of fluorescence-stained cells for various biological and biomedical tasks. The study investigated, compared, and enhanced four different methods for segmentation of cultured epithelial cells. The maximum-intensity linking (MIL) method, an improved MIL, a watershed method, and an improved watershed method based on morphological reconstruction were used. Three manually annotated datasets consisting of 261, 817, and 1,333 HeLa or L929 cells were used to compare the different algorithms. The comparisons and evaluations showed that the segmentation performance of methods based on the watershed transform was significantly superior to the performance of the MIL method. The results also indicate that using morphological opening by reconstruction can improve the segmentation of cells stained with a marker that exhibits the dotted surface of cells.

ENTROPY TOLERANT FUZZY C-MEANS IN MEDICAL IMAGES

Segmenting the Dynamic Contrast-Enhanced Breast Magnetic Resonance Images (DCE-BMRI) is an extremely important task to diagnose the disease because it has the highest specificity when acquired with high temporal and spatial resolution and is also corrupted by heavy noise, outliers, and other imaging artifacts. In this paper, we intend to develop efficient robust segmentation algorithms based on fuzzy clustering approach for segmenting the DCE-BMRs. Our proposed segmentation algorithms have been amalgamated with effective kernel-induced distance measure on standard fuzzy c-means algorithm along with the spatial neighborhood information, entropy term, and tolerance vector into a fuzzy clustering structure for segmenting the DCE-BMRI. The significant feature of our proposed algorithms is its capability to find the optimal membership grades and obtain effective cluster centers automatically by minimizing the proposed robust objective functions. Also, this article demonstrates the superiority of the proposed algorithms for segmenting DCE-BMRI in comparison with other recent kernel-based fuzzy c-means techniques. Finally the clustering accuracies of the proposed algorithms are validated by using silhouette method in comparison with existed fuzzy clustering algorithms.

Subspace learning for Mumford–Shah-model-based texture segmentation through texture patches

In this paper, we develop a robust and effective algorithm for texture segmentation and feature selection. The approach is to incorporate a patch-based subspace learning technique into the subspace Mumford-Shah (SMS) model to make the minimization of the SMS model robust and accurate. The proposed method is fully unsupervised in that it removes the need to specify training data, which is required by existing methods for the same model. We further propose a novel (to our knowledge) pairwise dissimilarity measure for pixels. Its novelty lies in the use of the relevance scores of the features of each pixel to improve its discriminating power. Some superior results are obtained compared to existing unsupervised algorithms, which do not use a subspace approach. This confirms the usefulness of the subspace approach and the proposed unsupervised algorithm.