Mean Shift Approach Research Articles

Segmentation of High‐Resolution Remote Sensing Images Using the Gabor Texture Feature‐Based Mean Shift Method

High‐resolution remote sensing images (HRRSIs) play an important role in the construction and development of society with their rich and detailed information. In the process of remote sensing image segmentation, the conventional method of mean shift usually involves spatial features and spectral features to preserve edge information and reduce the effect of noise. However, this traditional method often misses some important information in HRRSIs processing and reduces the segmentation accuracy. Thus, we proposed an improved mean shift image segmentation method based on Gabor texture features to make full use of the delicate information in HRRSIs. Specifically, the multiscale and multidirectional Gabor filters were employed in this study to extract pixel‐to‐pixel features from the QuickBird and GeoEye‐1 images, and the local variance was used to determine the optimal texture bandwidth. Two parameters, global segmentation (GS) quality index and error rate (ER), were adopted to evaluate the segmentation quality of these two HRRSIs. Results showed that the proposed approach in this study outperforms the traditional mean shift method on both remote sensing datasets. Compared with the traditional method, the improved mean shift approach corresponded to higher GS values and smaller ER values. This study confirms that the improved mean shift approach based on Gabor filtering has a potential for HRRSIs segmentation.

Membrane Cholesterol Prediction from Human Receptor using Rough Set based Mean-Shift Approach

In human physiology, cholesterol plays an imperative part in membrane cells which regulates the function of G-protein-coupled receptors (GPCR) family. Cholesterol is an individual type of lipid structure and about 90 percent of cellular cholesterol is present at plasma membrane region. Cholesterol Recognition/interaction Amino acid Consensus (CRAC) sequence, generally referred as the CRAC (L/V)-X1−5-(Y)-X1−5-(K/R) and the new cholesterol-binding domain is similar to the CRAC sequence, but exhibits the inverse orientation along the polypeptide chain i.e. CARC (K/R)-X1−5-(Y/F)-X1−5-(L/V). GPCR is treated as a biggest super family in human physiology and probably more than 900 protein genes included in this family. Among all membrane proteins GPCR is responsible for novel drug discovery in all pharmaceuticals industry. In earlier researches the researchers did not find the required number of valid motifs in terms of helices and motif types so they were lacking clinical relevance. The research gap here is that they were not able to predict the motifs effectively which are belonging to multiple motif types. To find out better motif sequences from human GPCR, we explored a hybrid computational model consisting of hybridization of Rough Set with Mean-Shift algorithm. In this paper we made comparison among our resulted output with other techniques such as fuzzy C-means (FCM), FCM with spectral clustering and we concluded that our proposed method targeted well on CRAC region in comparison to CARC region which have higher biological relevance in medicine industry and drug discovery.

Breast cancer segmentation based on modified Gaussian mean shift algorithm for infrared thermal images

ABSTRACT Accurate segmentation of infrared thermal images is a challenging issue in the breast cancer detection. This paper presents and evaluates novel segmentation method for breast cancer detection using infrared thermal images. The developed segmentation method uses developed segmentation algorithm that is hybrid of Gaussian Mean Shift (GMS) and roulette wheel selection approach. In the first stage of the developed segmentation method, the redundant portions of the infrared thermal image are removed and then, the infrared thermal image is divided into five and six clusters by applying the developed segmentation algorithm. The segmented infrared thermal image is created by multiplying the selected best image of each cluster. This image is used to diagnosis the normal breast from abnormal breast. The proposed method is developed using MATLAB 2017 and infrared thermal images of 64 patients. The results show that the average Dice similarity coefficient, Jaccard index and Hausdorff distance in the proposed segmentation method are 91.81%, 84.86%, and 4.87, respectively. The comparison results demonstrate that using the proposed segmentation method can improve the performance of the Computer-Aided-Detection (CAD) system compared to the CAD systems that use the Mean Shift (MS) and Fuzzy C-Means (FCM) segmentation algorithms.

Mean Shift versus Variance Inflation Approach for Outlier Detection—A Comparative Study

Outlier detection is one of the most important tasks in the analysis of measured quantities to ensure reliable results. In recent years, a variety of multi-sensor platforms has become available, which allow autonomous and continuous acquisition of large quantities of heterogeneous observations. Because the probability that such data sets contain outliers increases with the quantity of measured values, powerful methods are required to identify contaminated observations. In geodesy, the mean shift model (MS) is one of the most commonly used approaches for outlier detection. In addition to the MS model, there is an alternative approach with the model of variance inflation (VI). In this investigation the VI approach is derived in detail, truly maximizing the likelihood functions and examined for outlier detection of one or multiple outliers. In general, the variance inflation approach is non-linear, even if the null model is linear. Thus, an analytical solution does usually not exist, except in the case of repeated measurements. The test statistic is derived from the likelihood ratio (LR) of the models. The VI approach is compared with the MS model in terms of statistical power, identifiability of actual outliers, and numerical effort. The main purpose of this paper is to examine the performance of both approaches in order to derive recommendations for the practical application of outlier detection.

A distributed approximate nearest neighbors algorithm for efficient large scale mean shift clustering

Mean Shift clustering, as a generalization of the well-known k-means clustering, computes arbitrarily shaped clusters as defined as the basins of attraction to the local modes created by the density gradient ascent paths. Despite its potential for improved clustering accuracy, the Mean Shift approach is a computationally expensive method for unsupervised learning. We introduce two contributions aiming to provide approximate Mean Shift clustering, based on scalable procedures to compute the density gradient ascent and cluster labeling, with a linear time complexity, as opposed to the quadratic time complexity for the exact clustering. Both propositions are based on Locality Sensitive Hashing (LSH) to approximate nearest neighbors. When implemented on a serial system, these approximate methods can be used for moderate sized datasets. To facilitate the analysis of Big Data, a distributed implementation, written for the Spark/Scala ecosystem is proposed. An added benefit is that our proposed approximations of the density gradient ascent, when used as a pre-processing step in other clustering methods, can also improve the clustering accuracy of the latter. We present experimental results illustrating the effect of tuning parameters on cluster labeling accuracy and execution times, as well as the potential to solve concrete problems in Big Clustering.

Spatial and spectral regularization to discriminate tissues using multispectral photoacoustic imaging

Photoacoustics is a hybrid modality used to image biological tissues. As optical absorption of tissues depends on the wavelength of the transmitted light, multispectral photoacoustic datasets can be obtained by changing this wavelength. This study presents a regularization method to segment multispectral photoacoustic images based on both the spatial and spectral features of the dataset pixels. The proposed processing is adapted from the spatiotemporal mean-shift approach and cluster patterns with similar spectral profiles, i.e., the variation of the received amplitude among the wavelengths, independent of their initial position. The segmentation performance of this method has been experimentally tested on multispectral photoacoustic tomographic data. We initially used a phantom that contained fresh and stale liver samples, and then a second phantom that contained two blood dilutions or a colored absorber. Experimentally, a clustering performance greater than 98% is achieved. This method makes it possible to discriminate between different media, between the same medium as fresh or stale, and between the same medium with different dilutions.

Gray level co-occurrence matrix feature based object tracking in thermal infrared imagery

Object tracking plays a vital role in many computer vision systems applications, such as video surveillance, robotics, 3-D image reconstruction, medical imaging, human computer interface, etc. In many proposed approaches, feature-based object tracking is widely used due to its accuracy. Feature extraction and feature correspondence are two main components of feature-based object tracking. In our proposed method, we have used gray level co-occurrence matrix (GLCM) features for object tracking in thermal imagery. As spatial resolution of the thermal sensor is fairly coarse, it also implies that temperature scales are close but may not exactly be the same, which indicates the presence of mutually related pixels or group of pixels. The GLCM texture analysis is based on assumption that the texture information of an image is an average spatial relationship between the gray tones in the image. Thus, this similarity in spatial resolution properties makes GLCM features suitable for object tracking in thermal infrared imagery. Initially, the target blobs to be tracked are provided by object detection stage. Then, for a given target blob in a frame, we first calculate GLCM feature points and then find corresponding features in the next successive frame. The sum of squared differences between two feature point sets is calculated to find feature correspondence between two frames for object tracking. Simultaneously, the codebook of the center of the blobs for prediction of the target candidate region in the next frame is maintained in order to have robust object tracking under occlusions. GLCM-based object tracking in thermal imagery outperforms the color or LBP-based mean-shift approach. This algorithm is also able to track objects in split and merge condition. The accuracy of this algorithm also depends on the object detection stage, such as Kalman tracking.

WHEAT EAR DETECTION IN PLOTS BY SEGMENTING MOBILE LASER SCANNER DATA

Abstract. The use of Light Detection and Ranging (LiDAR) to study agricultural crop traits is becoming popular. Wheat plant traits such as crop height, biomass fractions and plant population are of interest to agronomists and biologists for the assessment of a genotype's performance in the environment. Among these performance indicators, plant population in the field is still widely estimated through manual counting which is a tedious and labour intensive task. The goal of this study is to explore the suitability of LiDAR observations to automate the counting process by the individual detection of wheat ears in the agricultural field. However, this is a challenging task owing to the random cropping pattern and noisy returns present in the point cloud. The goal is achieved by first segmenting the 3D point cloud followed by the classification of segments into ears and non-ears. In this study, two segmentation techniques: a) voxel-based segmentation and b) mean shift segmentation were adapted to suit the segmentation of plant point clouds. An ear classification strategy was developed to distinguish the ear segments from leaves and stems. Finally, the ears extracted by the automatic methods were compared with reference ear segments prepared by manual segmentation. Both the methods had an average detection rate of 85 %, aggregated over different flowering stages. The voxel-based approach performed well for late flowering stages (wheat crops aged 210 days or more) with a mean percentage accuracy of 94 % and takes less than 20 seconds to process 50,000 points with an average point density of 16 points/cm2. Meanwhile, the mean shift approach showed comparatively better counting accuracy of 95% for early flowering stage (crops aged below 225 days) and takes approximately 4 minutes to process 50,000 points.

Location-Based Parallel Tag Completion for Geo-Tagged Social Image Retrieval

Having benefited from tremendous growth of user-generated content, social annotated tags get higher importance in the organization and retrieval of large-scale image databases on Online Sharing Websites (OSW). To obtain high-quality tags from existing community contributed tags with missing information and noise, tag-based annotation or recommendation methods have been proposed for performance promotion of tag prediction. While images from OSW contain rich social attributes, they have not taken full advantage of rich social attributes and auxiliary information associated with social images to construct global information completion models. In this article, beyond the image-tag relation, we take full advantage of the ubiquitous GPS locations and image-user relationship to enhance the accuracy of tag prediction and improve the computational efficiency. For GPS locations, we define the popular geo-locations where people tend to take more images as Points of Interests (POI), which are discovered by mean shift approach. For image-user relationship, we integrate a localized prior constraint, expecting the completed tag sub-matrix in each POI to maintain consistency with users’ tagging behaviors. Based on these two key issues, we propose a unified tag matrix completion framework, which learns the image-tag relation within each POI. To solve the optimization problem, an efficient proximal sub-gradient descent algorithm is designed. The model optimization can be easily parallelized and distributed to learn the tag sub-matrix for each POI. Extensive experimental results reveal that the learned tag sub-matrix of each POI reflects the major trend of users’ tagging results with respect to different POIs and users, and the parallel learning process provides strong support for processing large-scale online image databases. To fit the response time requirement and storage limitations of Tag-based Image Retrieval (TBIR) on mobile devices, we introduce Asymmetric Locality Sensitive Hashing (ALSH) to reduce the time cost and meanwhile improve the efficiency of retrieval.

A Fuzzy Mean-Shift Approach to Lidar Waveform Decomposition

Waveform decomposition is a common step for exploitation of full-waveform lidar data. Much effort has been focused on designing algorithms based on the assumption that the returned waveforms follow a Gaussian mixture model where each component is a Gaussian. However, many real examples show that the waveform components can be neither Gaussian nor symmetric even when the emitted signal is Gaussian or symmetric. This paper proposes a nonparametric mixture model to represent lidar waveforms without any constraints on the shape of the waveform components. A fuzzy mean-shift algorithm is then developed to decompose the waveforms. This approach has the following properties: 1) It does not assume that the waveforms follow any parametric or functional distributions; 2) the waveform decomposition is treated as a fuzzy data clustering problem and the number of components is determined during the time of decomposition; and 3) neither peak selection nor noise floor filtering prior to the decomposition is needed. Experiments are conducted on a dataset collected over a dense forest area where significant skewed waveforms are demonstrated. As the result of the waveform decomposition, a highly dense point cloud is generated, followed by a subsequent filtering step to create a fine digital elevation model. Compared with the conventional expectation–maximization method, the fuzzy mean-shift approach yielded practically comparable and similar results. However, it is about three times faster and tends to lead to slightly fewer artifacts in the resultant digital elevation model.

Moving Object Tracking Based on Gaussian Kernel and Template Modelling

The Project presents object tracking from videos based on template matching using Gaussian kernel with probability distribution function and mean shift algorithm. Initially, the target will be selected from chosen video sequence to track desired object in consecutive frames. The target will be utilized to determine the probability distribution function for similarity measurement between target and current processing frames. Here, Gaussian kernel function and its gradient are used here to find the PDF for corresponding templates. Similarity between two different images will be measured by weighted sum of Gaussian coefficients and PDFS. Mean shift approach used here to shifting the starting coordinates of template to find its similar features in consecutive frames to detect desired objects. The dissimilarity between the target model and target candidates will be expressed by a metric derived from Bhattacharyya coefficient. The project simulated results shows that moving object from video will be tracked accurately at different position and shape with help of templates in a considerable amount of time.

Improved CRC Based Disparity Estimation of Vision System Using Local Adaptive Hue Census and Mean Shift Clustering

One of the key issues of stereo matching is the identification of corresponding pixels between stereo pairs. Several real time systems developed based on correlation models, still exhibit blurred borders and finer details distortions. This work proposes a new adaptive method which integrates hue based census transform with bilinear interpolation to attain accurate matching. Different weights were assigned to pixels in support window based on their color and spatial distances. Global mean shift approach was applied to obtain high confidence disparity map. Incorrect matches and enhances disparity were reduced with the help of cyclic redundancy check approach. Finally, triangulation was used to generate an effective 3D view. Simulation results generated a better disparity (nearly 85 %) with minimum mismatches than existing approaches.

An OPMA for Robust Mutual Coupling Coefficients Estimation of URA with Single Snapshot in MIMO HF Sky-Wave Radar

Due to the fluctuation of the signal-to-noise ratio (SNR) and the single snapshot case in the MIMO HF sky-wave radar system, the accuracy of the online estimation of the mutual coupling coefficients matrix of the uniform rectangle array (URA) might be degraded by the classical approach, especially in the case of low SNR. In this paper, an Online Particle Mean-Shift Approach (OPMA) is proposed, which is to get a relatively more effective estimation of the mutual coupling coefficients matrix with the low SNR. Firstly, the spatial smoothing technique combined with the MUSIC algorithm of URA is introduced for the DOA estimation of the multiple targets in the case of single snapshot which are taken as coherent sources. Then, based on the idea of the particle filter, the online particles with a moderate computational complexity are used to generate some different estimation results. Finally, the mean-shift algorithm is applied to get a more robust estimate of the equivalent mutual coupling coefficients matrix. The simulation results demonstrate the validity of the proposed approach in terms of the success probability, the statistics of bias, and the variance. The proposed approach is more robust and more accurate than the other two approaches.

Efficient and Robust Learning for Sustainable and Reacquisition-Enabled Hand Tracking.

The use of machine learning approaches for long-term hand tracking poses some major challenges such as attaining robustness to inconsistencies in lighting, scale and object appearances, background clutter, and total object occlusion/disappearance. To address these issues in this paper, we present a robust machine learning approach based on enhanced particle filter trackers. The inherent drawbacks associated with the particle filter approach, i.e., sample degeneration and sample impoverishment, are minimized by infusing the particle filter with the mean shift approach. Moreover, to instill our tracker with reacquisition ability, we propose a rotation invariant and efficient detection framework named beta histograms of oriented gradients. Our robust appearance model operates on the red, green, blue color histogram and our newly proposed rotation invariant noise compensated local binary patterns descriptor, which is a noise compensated, rotation invariant version of the local binary patterns descriptor. Through our experiments, we demonstrate that our proposed hand tracker performs favorably against state-of-the-art algorithms on numerous challenging video sequences of hand postures, and overcomes the largely unsolved problem of redetecting hands after they vanish and reappear into the frame.

English

Colossal amount of research has been done in creating many different approaches and algorithms for medical image segmentation, but it is still complicated to evaluate all the images. However the problem remains challenging, with no general and unique solution in computer-aided diagnosis. This paper provides medical image segmentation based on Clustering for computer tomography images. In this paper, we consider a mean shift segmentation and medoid shift segmentation method. We validate the mean shift and medoid shift medical image segmentation approach with the parameters in terms of sensitivity, specificity and accuracy. The Real time dataset is used to evaluate the performance of the proposed method. The experimental result shows that the medoid shift segmentation method gives more accurate and robust segmentation results than mean shift segmentation method.

Multiclass Image Segmentation Based on Pixel and Segment Level

Multi-class image segmentation (or pixel labeling) is one of the most important and challenging tasks in computer vision. Currently, many different methods for this task can be broadly categorized into two types according to their choice of the partitioning of the image space, i.e., pixels or segments. However, each choice of the two types of methods comes with its share of advantages and disadvantages. In this study, we construct a novel CRF model to integrate features extracted from pixel and segment levels. We exploit segments generated by Constrained Parametric Min Cuts (CPMC) algorithm in the proposed framework, instead of commonly used unsupervised segmentation method (e.g., mean-shift approach). Additionally, the recognition based on these segments is also integrated into the model, which possible corrects classification mistakes caused by the unary term based on information derived from pixel level. We experimentally demonstrate our model’s quantitative and qualitative improvements over the baseline methods.

Remote sensing of seasonal variability of fractional vegetation cover and its object-based spatial pattern analysis over mountain areas

Fractional vegetation cover (FVC) is an important indicator of mountain ecosystem status. A study on the seasonal changes of FVC can be beneficial for regional eco-environmental security, which contributes to the assessment of mountain ecosystem recovery and supports mountain forest planning and landscape reconstruction around megacities, for example, Beijing, China. Remote sensing has been demonstrated to be one of the most powerful and feasible tools for the investigation of mountain vegetation. However, topographic and atmospheric effects can produce enormous errors in the quantitative retrieval of FVC data from satellite images of mountainous areas. Moreover, the most commonly used analysis approach for assessing FVC seasonal fluctuations is based on per-pixel analysis regardless of the spatial context, which results in pixel-based FVC values that are feasible for landscape and ecosystem applications. To solve these problems, we proposed a new method that incorporates the use of a revised physically based (RPB) model to correct both atmospheric and terrain-caused illumination effects on Landsat images, an improved vegetation index (VI)-based technique for estimating the FVC, and an adaptive mean shift approach for object-based FVC segmentation. An array of metrics for segmented FVC analyses, including a variety of area metrics, patch metrics, shape metrics and diversity metrics, was generated. On the basis of the individual segmented FVC values and landscape metrics from multiple images of different dates, remote sensing of the seasonal variability of FVC was conducted over the mountainous area of Beijing, China. The experimental results indicate that (a) the mean value of the RPB–NDVI in all seasons was increased by approximately 10% compared with that of the atmospheric correction-NDVI; (b) a strong consistency was demonstrated between ground-based FVC observations and FVC estimated through remote sensing technology (R2=0.8527, RMSE=0.0851); and (c) seasonal changes in the landscape characteristics existed, and the landscape diversity reached its maximum in May and June in the study area.

Dynamics of a mean-shift-like algorithm and its applications on clustering

The Mean-Shift (MS) algorithm and its variants have wide applications in pattern recognition and computer vision tasks such as clustering, segmentation, and tracking. In this paper, we study the dynamics of the algorithm with Gaussian kernels, based on a Generalized MS (GMS) model that includes the standard MS as a special case. First, we prove that the GMS has solutions in the convex hull of the given data points. By the principle of contraction mapping, a sufficient condition, dependent on a parameter introduced into Gaussian kernels, is provided to guarantee the uniqueness of the solution. It is shown that the solution is also globally stable and exponentially convergent under the condition. When the condition does not hold, the GMS algorithm can possibly have multiple equilibriums, which can be used for clustering as each equilibrium has its own attractive basin. Based on this, the condition can be used to estimate an appropriate parameter which ensures the GMS algorithm to have its equilibriums suitable for clustering. Examples are given to illustrate the correctness of the condition. It is also shown that the use of the multiple-equilibrium property for clustering, on the data sets such as IRIS, leads to a lower error rate than the standard MS approach, and the K-Means and Fuzzy C-Means algorithms.

CNVnator: An approach to discover, genotype, and characterize typical and atypical CNVs from family and population genome sequencing

Copy number variation (CNV) in the genome is a complex phenomenon, and not completely understood. We have developed a method, CNVnator, for CNV discovery and genotyping from read-depth (RD) analysis of personal genome sequencing. Our method is based on combining the established mean-shift approach with additional refinements (multiple-bandwidth partitioning and GC correction) to broaden the range of discovered CNVs. We calibrated CNVnator using the extensive validation performed by the 1000 Genomes Project. Because of this, we could use CNVnator for CNV discovery and genotyping in a population and characterization of atypical CNVs, such as de novo and multi-allelic events. Overall, for CNVs accessible by RD, CNVnator has high sensitivity (86%-96%), low false-discovery rate (3%-20%), high genotyping accuracy (93%-95%), and high resolution in breakpoint discovery (<200 bp in 90% of cases with high sequencing coverage). Furthermore, CNVnator is complementary in a straightforward way to split-read and read-pair approaches: It misses CNVs created by retrotransposable elements, but more than half of the validated CNVs that it identifies are not detected by split-read or read-pair. By genotyping CNVs in the CEPH, Yoruba, and Chinese-Japanese populations, we estimated that at least 11% of all CNV loci involve complex, multi-allelic events, a considerably higher estimate than reported earlier. Moreover, among these events, we observed cases with allele distribution strongly deviating from Hardy-Weinberg equilibrium, possibly implying selection on certain complex loci. Finally, by combining discovery and genotyping, we identified six potential de novo CNVs in two family trios.

Fast and Robust Moving Objects Detection based on Non-parametric Background Modeling

Fast and reliable detection of moving objects is one of the important requirements for many computer vision and video analysis applications. Mean shift based non-parametric background modeling supports more sensitive and robust detection in dynamic outdoor scenes. However it is prohibitive to real-time applications such as video surveillance. This paper aims to deal with the limitation of high computational complexity. Firstly, coarse to fine methods are proposed to avoid raster scanning entire image. Foreground pixels are detected in coarse level to roughly locate the foreground objects in the image, and then fine detection is performed on the corresponding blocks gradually. Secondly ， fast mean shift approach is presented according to temporal dependencies. Mean shift iterations are performed starting from incoming data and the modes obtained last time. The experimental results show that the proposed algorithm is effective and efficient in dynamic environment. The proposed algorithm has been applied to move objects detection in our real-time marine video surveillance system.