Image Segmentation Step Research Articles

Recognition Algorithm and Optimization Experiments on Tomato Picking Robots

In order to improve the recognition accuracy of vision system on tomato picking robots, the paper proposed a method of feature extraction and recognition for ripe tomato based on illumination irrelevant images and support vector machine (SVM). In this method, we adopted vector median filter (VMF) to process the tomato images to eliminate noise and make the images more smooth firstly. To avoid the effects of natural environment illumination to the vision system, we processed tomato images and obtained the tomato illumination irrelevant images according to color constancy algorithm of the single pixel. Secondly, we segmented illumination irrelevant images using OSTU method, separated multiple objects by a watershed algorithm based on distance transform and got the target area with mathematical morphology. Also we extracted color, shape and textural features of the ripe tomatoes. Finally, we did experiments on recognizing tomatoes using support vector machine (SVM) with different kernel functions. At the same time, in order to obtain optimal model of SVM, we adopted cross validation and grid search method to optimize the model parameters. The experiment results show that illumination irrelevant processing not only can eliminate the influence of light intensity, but save a gray transferring step for further image segmentation. SVM with radial basis function is better than other kernel functions SVM and the tomato recognition accuracy is 95.7%. Through optimizing the parameter C and r of radial basis function, the tomato recognition accuracy reaches up to 96.9% with the increase of 1.2% when C and r is 4 and 16 respectively. This proves that it's feasible and effective to optimize SVM's parameters by cross validation and grid search method, which provide foundation for further research on vision system of tomato picking robots.

Grayscale Conversion of Histopathological Slide Images as a Preprocessing Step for Image Segmentation

This paper presents a method of conversion of color image of histopathological slides to grayscale as a preprocessing step for image segmentation. The method utilizes characteristics of these images to reduce the amount of information lost during the conversion. It is a global mapping scheme and the same color does not map to two different grayscale values. Color information is added to the brightness information to reduce information loss and the weights for the added color are dependent on the distribution of the pixels in a particular image. The method was tested on a large set of histopathological slide images and convincing results were obtained some of which has been published in this paper.

Psychological Interpretation of Drawings by Color Analysis for Mental Therapy

In order to provide psychotherapists with an objective criterion for evaluating client's mental states, this paper proposes a method for analyzing a drawing that a client makes in drawing therapy for four psychological primary colors. The method consists of two steps: an image segmentation step and a color analysis step. In the first step, those areas which seem what the client intended to depict are extracted from the entire digitized image. The effective resolution in the extraction is made variable with the values of parameters in mathematically morphological processing. In the second step, the pixels of the extracted areas are classified into four categories, corresponding to the psychological primary colors, as determined by the dominant component of a pixel color. A case study of employing the color analyses in drawing therapy was carried out with a series of drawings which were obtained in an actual clinical therapy. A comparison of the objective results of the analysis and subjective evaluations obtained in the process of cognitive behavioral therapy has demonstrated that the analyses are useful to describe a variation in a client's mental state in terms of psychological properties of colors. In conclusion, the color analysis method in drawing therapy works in cooperation with cognitive behavioral therapy.

IMAGE SEGMENTATION METHODS FOR BRAIN MRI IMAGES

In Image Processing, extracting the region of interest is a very challenging task. To extract information, pre-processing algorithms are important in MRI image. Edge detection is a task in which points in image are identified at which brightness changes sharply or it has discontinuities. It is an essential pre- processing step in medical image segmentation, for object recognition of the human organs. The applications of medical image segmentation are 3D reconstruction and quantitative analysis and so on. We used MRI images because MRI images give best view of tissues in any part of human body. In this paper, difficulties of edge detection in brain magnetic resonance images are considered and a new approach to edge detection is introduced. There are many traditional edge detection methods for extracting edges from images have been introduced such as gradient based operators like sobel, prewitt, robert were initially used for edge detection, but they did not give sharp edges and were highly sensitive to noise image. And in medical field accuracy is important fact. To overcome these difficulties, we proposed new method called as Active Contour method or snake model. . In the field of medical segmentation, Active contour method is one of popular research topic. This method is used for detecting brain region based on their energy function. In order to compare between them, one slice of MRI image tested with these methods. The traditional and proposed edge detection algorithms are implemented in MATLAB and results of proposed method are presented and compared with traditional approach.

An improved K-means clustering method for cDNA microarray image segmentation.

Microarray technology is a powerful tool for human genetic research and other biomedical applications. Numerous improvements to the standard K-means algorithm have been carried out to complete the image segmentation step. However, most of the previous studies classify the image into two clusters. In this paper, we propose a novel K-means algorithm, which first classifies the image into three clusters, and then one of the three clusters is divided as the background region and the other two clusters, as the foreground region. The proposed method was evaluated on six different data sets. The analyses of accuracy, efficiency, expression values, special gene spots, and noise images demonstrate the effectiveness of our method in improving the segmentation quality.

Semiautomatic segmentation of longitudinal computed tomography images in a rat model of lung injury by surfactant depletion.

Quantitative analysis of computed tomography (CT) is essential to the study of acute lung injury. However, quantitative CT is made difficult by poor lung aeration, which complicates the critical step of image segmentation. To overcome this obstacle, this study sought to develop and validate a semiautomated, multilandmark, registration-based scheme for lung segmentation that is effective in conditions of poor aeration. Expiratory and inspiratory CT images were obtained in rats (n = 8) with surfactant depletion of incremental severity to mimic worsening aeration. Trained operators manually delineated the images to provide a comparative landmark. Semiautomatic segmentation originated from a single, previously segmented reference image obtained at healthy baseline. Deformable registration of the target images (after surfactant depletion) was performed using the symmetric diffeomorphic transformation model with B-spline regularization. Registration used multiple landmarks (i.e., rib cage, spine, and lung parenchyma) to minimize the effect of poor aeration. Then target images were automatically segmented by applying the calculated transformation function to the reference image contour. Semiautomatically and manually segmented contours proved to be highly similar in all aeration conditions, including those characterized by more severe surfactant depletion and expiration. The Dice similarity coefficient was over 0.9 in most conditions, confirming high agreement, irrespective of poor aeration. Furthermore, CT density-based measurements of gas volume, tissue mass, and lung aeration distribution were minimally affected by the method of segmentation. Moving forward, multilandmark registration has the potential to streamline quantitative CT analysis by enabling semiautomatic image segmentation of lungs with a broad range of injury severity.

An active contour model based on fused texture features for image segmentation

Texture image segmentation plays an important role in various computer vision tasks. In this paper, a convex texture image segmentation model is proposed. First, the texture features of Gabor and GLCM (gray level co-occurrence matrix) are extracted for original image. Then, the two kinds of texture features are fused together to effectively construct a discriminative feature space by concatenating with each other. In the image segmentation step, a convex energy function is defined by taking the non-convex vector-valued model of Active Contour without Edges (ACWE) into a global minimization framework (GMAC). The proposed global minimization energy function with fused textures (GMFT) can avoid the existence of local minima in the minimization of the vector-valued ACWE model. In addition, a fast dual formulation is adopted to achieve the efficient contour evolution. The experimental results on synthetic and natural animal images demonstrate that the proposed GMFT model obtains more satisfactory segmentation results compared to two state-of-the-art methods in terms of segmentation accuracy and efficiency.

Sealed-Crack Detection Algorithm Using Heuristic Thresholding Approach

AbstractSealed cracks are frequently observed on the road pavement due to pavement crack sealing. A sealed crack is a type of road surface distress, whose assessment is essential in pavement management system. In this paper, the state-of-the-art in automated pavement surface crack detection is presented. An algorithm is also developed to automatically detect sealed cracks in pavement surface images. The algorithm steps are elaborated with case study samples. Moreover, in image segmentation step, a method is developed based on a local minimum approach. The proposed algorithm is tested through various pavement surface images collected in an exhaustive case study. The evaluation is conducted with respect to the performance and robustness perspectives. The experimental results indicate that the developed algorithm has high accuracy and can consistently detect sealed cracks in different environments.

A Hybridized Disparity Computation Fusing Disparity Space Image and Multi Resolution Image Segmentation

D information is getting more importance nowadays.There are several ways to get 3D information.One way to get 3D information involves the laser based depth estimators, which is a very costly process. Other way is to extract 3D information from 2D stereo image pairs by using stereo matching techniques. Stereo matching is a vast area of research. It can be classified into area (window) based approach, feature based approach, and optimization based approach. Area based approach generally generates dense disparity map with low accuracy and low computation time. Feature based method produces highly accurate and sparse disparity map with low computation time.Optimization based method produces dense disparity map with high accuracy and high execution time.By keeping these things in mind, we proposed a new hybrid way for disparity computation. The proposed method consists of three steps. Theyare disparityestimation, image segmentation, and disparity refinement. Since the ultimate goal of stereo matching is to obtain dense disparity map with high accuracy and low execution time, we select optimization based approach for disparity estimation step and for image segmentation step we select multi resolution image segmentation. At the end, disparity refinement is done by combining the result of both the previous steps.As there are several optimization techniques, we choose disparity space image (DSI) baseddynamic programming (DP).We tested the proposed method on several stereo pairs and found that method produced reasonably good quality results. Index Terms—Stereo matching, DSI, DP.

Object-based change detection in wind storm-damaged forest using high-resolution multispectral images

Natural disasters are generally brutal and may affect large areas, which then need to be rapidly mapped to assess the impacts of such events on ecosystems and to prevent related risks. Ground investigations may be complex, whereas remote-sensing techniques enable a fast regional-scale assessment of damage and offer a cost-effective option for large and inaccessible areas. Here, an efficient, quasi-automatic object-based method for change mapping using high-spatial-resolution (HR) (5–10 m) satellite imagery is proposed. Our contribution comprises two main novelties with respect to similar works in forestry. First, an automatic feature selection process optimizes the image segmentation and classification steps via an original calibration-like procedure. Second, an automatic bitemporal classification enables the separation of damaged and intact areas thanks to a new descriptor based on the level of fragmentation of the obtained regions. The mean shift algorithm is used in both the segmentation and classification processes. The method was assessed in a maritime pine forest using bitemporal HR Formosat-2 multispectral images acquired pre- and post-Windstorm Klaus, which occurred in January 2009 in southwestern France. The binary overall classification accuracy reached 87.8% and outperformed a pixel-based K-means classification with no feature selection. A thematic analysis of the results highlights the correlation between the ages of trees and their sensitivity to wind.

基于语义约束与Graph Cuts的稠密三维场景重建

A new and effective dense 3D scene reconstruction algorithm is proposed in this work. At first, in order to speedup the reconstruction process and reduce possible unnecessary interference by unrelated regions, such as sky, ground, etc., to the desired architectural scene, a semantic image segmentation step is carried out at first to mask those uninterested parts. After that, an initial depth map is obtained by Daisy-feature based 3D point propagation. To tackle the difficult dense reconstruction problem of poorly textured regions and slanted surfaces, the problem is formulated as a plane inference problem under the MRF energy minimization framework to assign a possible planar patch to each image superpixel. More specifically, at first, the plane principal directions and possible non-principal are clustered by the affinity propagation from the reconstructed point clouds, then the plane inference is carried out along the principal directions first, followed by the non-principal directions. In each of the above two plane inference steps, a Graph Cuts based algorithm is employed to minimize a properly designed cost function which takes into account a variety of factors, such as appearance similarity, the constraints from the initial depth map, spatial planes priors, as well as geometric relations between spatial planes etc. Experiments on the standard data set as well as our own data set show that the our proposed algorithm can satisfactorily handle many challenging factors, e.g. illumination variation, perspective distortion, poorly textured regions, etc., to efficiently recover complete dense depth map of buildings.

Representative elementary volume assessment of three-dimensional x-ray microtomography images of heterogeneous materials: Application to limestones

Over the last 15 years, x-ray microtomography has become a useful technique to obtain morphological, structural, and topological information on materials. Moreover, these three-dimensional (3D) images can be used as input data to assess certain properties (e.g., permeability) or to simulate phenomena (e.g., transfer properties). In order to capture all the features of interest, high spatial resolution is required. This involves imaging small samples, raising the question of the representativity of the data sets. In this article, we (i) present a methodology to analyze the microstructural properties of complex porous media from 3D images, (ii) assess statistical representative elementary volumes (REVs) for such materials; and (iii) establish criteria to delimit these REVs. In the context of cultural heritage conservation, a statistical study was done on 30 quarry samples for three sorts of stones. We first present the principles of x-ray microtomography experiments and emphasize the care that must be taken in the 3D image segmentation steps. Results show that statistical REVs exist for these media and are reached for the image sizes studied (1300 × 1300 × 1000 voxels) for two characteristics: porosity and chord length distributions. Furthermore, the estimators used (porosity, autocorrelation function, and chord length distributions) are sufficiently sensitive to quantitatively distinguish these three porous media from each other. Lastly, this study puts forward criteria based on the above-mentioned estimators to evaluate the REVs. These criteria avoid having to repeat the statistical study for each new material studied. This is particularly relevant to quantitatively monitor the modifications in materials (weathering, deformation …) or to determine the smallest 3D volume for simulation in order to reduce computing time.

Ruler Detection for Autoscaling Forensic Images

The assessment of forensic photographs often requires the calibration of the resolution of the image so that accurate measurements can be taken of crime-scene exhibits or latent marks. In the case of latent marks, such as fingerprints, image calibration to a given dots-per-inch is a necessary step for image segmentation, preprocessing, extraction of feature minutiae and subsequent fingerprint matching. To enable scaling, such photographs are taken with forensic rulers in the frame so that image pixel distances can be converted to standard measurement units (metric or imperial). In forensic bureaus, this is commonly achieved by manual selection of two or more points on the ruler within the image, and entering the units of the measure distance. The process can be laborious and inaccurate, especially when the ruler graduations are indistinct because of poor contrast, noise or insufficient resolution. Here the authors present a fully automated method for detecting and estimating the direction and graduation spacing of rulers in forensic photographs. The method detects the location of the ruler in the image and then uses spectral analysis to estimate the direction and wavelength of the ruler graduations. The authors detail the steps of the algorithm and demonstrate the accuracy of the estimation on both a calibrated set of test images and a wide collection of good and poor quality crime-scene images. The method is shown to be fast and accurate and has wider application in other imaging disciplines, such as radiography, archaeology and surveying.

A Novel Algorithm for Edge Detection of Noisy Medical Images

Medical image edge detection is an important work for object recognition of the human organs, and it is an essential pre-processing step in medical image segmentation and 3D reconstruction. Although many edge-detection evaluation methods have been developed in the past years, however this is still a challenging and unsolved problem. Conventionally, edge is detected according to some early brought forward algorithms like Canny, LOG, Sobel, Prewitt, Roberts algorithms but in theory they belong to the high pass filtering, which are not fit for noise medical image edge detection because noise and edge belong to the scope of high frequency. In real world applications, medical images contain object boundaries and object shadows and noise. Therefore, they may be difficult to distinguish the exact edge from noise or trivial geometric features. After studying all traditional methods of edge detection, it has been analyzed that for these situations, a new algorithm is needed which is optimal. In this paper, we propose a new algorithm for edge detection of noisy medical images based on both Tsallis and Shannon entropy together. The performance of our method is compared against other methods by using blood cells image corrupted with various levels of salt and pepper. It is observed that the proposed algorithm displayed superior noise resilience and decrease the computation time.

User assisted disparity remapping for stereo images

This study concentrates on user assisted disparity remapping for stereo image footage, i.e. the disparity of an object of interest is altered while leaving the remaining scene unattended. This application is useful in the sense that it provides a method for emphasizing/de-emphasizing an object on the scene by adjusting its depth with respect to the camera. The proposed technique can also be used as a post-processing step for retargeting stereoscopic footage on different display sizes and resolutions. The proposed technique involves an MRF-based energy minimization step for interactive stereo image segmentation, for which user assistance on only one of the stereo pairs is required for determining the location of stereo object pair. A key contribution of the proposed study is elimination of dense disparity estimation step from the pipeline. This step is realized through a sparse feature matching technique between the stereo pairs. Moreover, by the help of the proposed technique, novel disparity adjusted views are synthesized using the produced stereo object segments and background information for the images. Qualitative and quantitative evaluation of the generated segments and the disparity adjusted images prove the functionality and superiority of the proposed technique.

Image Segmentations for Through-the-Wall Radar Target Detection

Detection of stationary targets using pixel-wise likelihood ratio test (LRT) detectors has been recently proposed for through-the-wall radar imaging (TWRI) applications. We employ image segmentation techniques, in lieu of LRT, for target detection in TWRI. More specifically the widely used between-class variance (BCV) thresholding, maximum entropy segmentation, and K-means clustering are considered to aid in removing the clutter, resulting in enhanced radar images with target regions only. For the case when multiple images of the same scene are available through diversity in polarization and/or vantage points around a building structure, we propose to use image fusion, following the image segmentation step, to generate an enhanced composite image. In particular, additive, multiplicative, and fuzzy logic fusion techniques are considered. The performance of the segmentation and fusion schemes is evaluated and compared with that of the assumed LRT detector using both electromagnetic (EM) modeling and real data collected in a laboratory environment. The results show that, although the principles of segmentation and detection are different, the segmentation techniques provide either comparable or improved performance over the LRT detector. Specifically, in the cases considered, the maximum entropy segmentation produces the best results for detection of targets inside building structures. For fusion of multiple segmented images of the same scene, the fuzzy logic fusion outperforms the other methods.

A Hybrid Approach to Parallel Connected Component Labeling Using CUDA

Connected component labeling (CCL) is a mandatory step in image segmentation where each object in an image is identified and uniquely labeled. Sequential CCL is a time-consuming operation and thus is often implemented within parallel processing framework to reduce execution time. Several parallel CCL methods have been proposed in the literature. Among them are NSZ label equivalence (NSZ- LE) method and modified 8 directional label selection (M8DLS) method. It was shown that M8DLS outperforms NSZ-LE and M8DLS is by far the best. In this paper we propose a new parallel CCL algorithm termed as HYBRID1 that hybridizes M8DLS and Kernel C method with some modification and show that it runs faster than M8DLS for various kinds of images.  Index Terms—connected component labeling, CUDA, GPU, parallel

Medical Images Edge Detection Based on Mathematical Morphology

Medical images edge detection is an important work for object recognition of the human organs and it is an important pre-processing step in medical image segmentation and reconstruction. Conventionally, edge is detected according to gradient-based algorithm and template-based algorithm, but they are not so good for noise medical image edge detection. In this paper, basic mathematical morphological theory and operations are introduced, and then a novel mathematical morphological edge detection algorithm is proposed to detect the edge of medical images with salt-and-pepper noise. The simulation results shows that the novel mathematical morphological edge detection algorithm is more efficient for image denoising and edge detection than the usually used template-based edge detection algorithms and general morphological edge detection algorithms. It has been observed that the proposed morphological edge detection algorithm performs better than sobel, prewitt, roberts and canny’s edge detection algorithm. In this paper the comparative analysis of various image edge detection techniques is presented using MATLAB 8.0 .

Color texture analysis based on fractal descriptors

Color texture classification is an important step in image segmentation and recognition. The color information is especially important in textures of natural scenes, such as leaves surfaces, terrains models, etc. In this paper, we propose a novel approach based on the fractal dimension for color texture analysis. The proposed approach investigates the complexity in R, G and B color channels to characterize a texture sample. We also propose to study all channels in combination, taking into consideration the correlations between them. Both these approaches use the volumetric version of the Bouligand-Minkowski Fractal Dimension method. The results show a advantage of the proposed method over other color texture analysis methods.

An automated approach for segmenting and classifying a large sample of multi-date Landsat imagery for pan-tropical forest monitoring

The TREES-3 project of the Joint Research Centre aims at assessing tropical forest cover changes for the periods 1990–2000 and 2000–2010 using a sample-based approach. This paper refers to the 1990–2000 assessment. Extracts of Landsat satellite imagery (20 km × 20 km) are analyzed for these reference dates for more than 4000 sample sites distributed systematically across the tropical belt. For the processing and analysis of such a large amount of satellite imagery a robust methodological approach for forest mapping and change detection has been developed. This approach comprises two automated steps of multi-date image segmentation and object-based land cover classification (based on a supervised spectral library), followed by an intense phase of visual control and expert refinement. Image segmentation is done at two spatial scales, introducing the concept of a minimum mapping unit via the automated selection of a site-specific scale parameter. The automated segmentation of land cover polygons and the pre-classification of land cover types mainly aim at avoiding manual delineation and at reducing the efforts of visual interpretation of land cover to a reasonable level, making the analysis of 4000 sample sites feasible. The importance of visual control and correction can be perceived when comparing to the initial automatic classification result: about 20% of the polygon labels were changed through expert knowledge by visual interpretation. The component of visual refinement of the mapping results had also a notable impact on forest area and change estimates: for a set of sample sites in Southeast Asia (~ 90% of all sites of SE-Asia) the rate of change in tree cover (deforestation) was assessed at 0.9% and 1.6% before and after visual control, respectively.