Multiband Image Segmentation Research Articles

Binary Segmentation of Multiband Images

We present a method for binary segmentation of multiband images based on a combination of dimensionality reduction techniques (Weighted PCA and Quadratic Programming Feature Selection), classi- cation methods (Gaussian Mixtures Models and Random Forest) and segmentation method (Quadratic Markov Measure Field Models). In this work, four pixels descriptors are presented: Color, Discrete Cosine Trans- form, Gradient Fields and Adjacency Matrix. Our method combines the outcome of several classiers using an optimization criterion. That results in a robust method for image segmentation based on color, textures and orientation. We evaluate our method capabilities with dierent image types for example: color images in RGB format and satellite images. Experimental results demonstrated our method performance.

Improving Parameters Selection of a Seeded Region Growing Method for Multiband Image Segmentation

In the last decade, Object Based Image Analysis (OBIA) has been accepted as an effective method for processing high spatial resolution multiband images. This image analysis method is an approach that starts with the segmentation of the image. Image segmentation in general is a procedure to partition an image into homogenous groups (segments). In practice, visual interpretation is often used to assess the quality of segmentation and the analysis relies on the experience of an analyst. In an effort to address the issue, in this study, we evaluate several seed selection strategies for an automatic image segmentation methodology based on a seeded region growing-merging approach. In order to evaluate the segmentation quality, segments were subjected to spatial autocorrelation analysis using Moran's I index and intra-segment variance analysis. We apply the algorithm to image segmentation using an aerial multiband image.

A Modified Level Set Approach for Segmentation of Multiband Polarimetric SAR Images

This paper investigates the application of a level set method for the automated multiphase segmentation of multiband and polarimetric synthetic aperture radar (SAR) images. The level set formulation is used to form an energy functional that includes the image statistical information defined on active contours. In addition to the classical Wishart/Gaussian distribution for locating region boundaries, edge information is incorporated into the energy functional to improve the performance of polarimetric data segmentation. An active contour model with an edge indicator is proposed by assuming that the image boundary term follows a Gibbs prior. An empirical parameter setting criterion is developed to ensure that the components of the energy functional are in proper proportion. We then investigate the multiphase extension for energy minimization, and we use a piecewise constant model to embed the proposed active contour model. Synthetic and real multiband polarimetric SAR data are used for verification. The experiments show that our method is superior to another level set method based on the Wishart/Gaussian distribution, in which SAR edge information is not included, particularly for discriminating among low-contrast regions. Furthermore, results also show that segmentation is improved when multiband data are used in the level set framework.

Multiband Image Segmentation and Object Recognition for Understanding Road Scenes

This paper presents a novel method for semantic segmentation and object recognition in a road scene using a hierarchical bag-of-textons method. Current driving-assistance systems rely on multiple vehicle-mounted cameras to perceive the road environment. The proposed method relies on integrated color and near-infrared images and uses the hierarchical bag-of-textons method to recognize the spatial configuration of objects and extract contextual information from the background. The histogram of the hierarchical bag-of-textons is concatenated to textons extracted from a multiscale grid window to automatically learn the spatial context for semantic segmentation. Experimental results show that the proposed method has better segmentation accuracy than the conventional bag-of-textons method. By integrating it with other scene interpretation systems, the proposed system can be used to understand road scenes for vehicle environment perception.

Approach to nonparametric cooperative multiband segmentation with adaptive threshold

We present a new nonparametric cooperative approach to multiband image segmentation. It is based on cooperation between region-growing segmentation and edge segmentation. This approach requires no input data other than the images to be processed. It uses a spectral homogeneity criterion whose threshold is determined automatically. The threshold is adaptive and varies depending on the objects to be segmented. Applying this new approach to very high resolution satellite imagery has yielded satisfactory results. The approach demonstrated its performance on images of varied complexity and was able to detect objects of great spatial and spectral heterogeneity.

Bayesian inference for multiband image segmentation via model-based cluster trees

We consider the problem of multiband image clustering and segmentation. We propose a new methodology for doing this, called model-based cluster trees. This is grounded in model-based clustering, which bases inference on finite mixture models estimated by maximum likelihood using the EM algorithm, and automatically chooses the number of clusters by Bayesian model selection, approximated using BIC, the Bayesian Information Criterion. For segmentation, model-based clustering is based on a Markov spatial dependence model. In the Markov model case, the Bayesian model selection criterion takes account of spatial neighborhood information, and is termed PLIC, the Pseudolikelihood Information Criterion. We build a cluster tree by first segmenting an image band, then using the second band to cluster each of the level 1 clusters, and continuing if required for further bands. The tree is pruned automatically as a part of the algorithm by using Bayesian model selection to choose the number of clusters at each stage. An efficient algorithm for implementing the methodology is proposed. An example is used to evaluate this new approach, and the advantages and disadvantages of alternative approaches to multiband segmentation and clustering are discussed.

Color image segmentation using histogram multithresholding and fusion

Abstract A novel method for multiband image segmentation has been proposed. The method is based on segmentation of subsets of bands using multithresholding followed by the fusion of the resulting segmentation “channels”. For color images the band subsets are chosen as the RB, RG and BG pairs, whose two-dimensional histograms are processed via a peak-picking algorithm to effect multithresholding. The segmentation maps are first fused by running a label concordance algorithm and then smoothed by a spatial–chromatic majority filter. It is shown that for multiband images, multithresholding subsets of bands followed by a fusion stage results in improved performance and running time.

Region competition: unifying snakes, region growing, and Bayes/MDL for multiband image segmentation

We present a novel statistical and variational approach to image segmentation based on a new algorithm, named region competition. This algorithm is derived by minimizing a generalized Bayes/minimum description length (MDL) criterion using the variational principle. The algorithm is guaranteed to converge to a local minimum and combines aspects of snakes/balloons and region growing. The classic snakes/balloons and region growing algorithms can be directly derived from our approach. We provide theoretical analysis of region competition including accuracy of boundary location, criteria for initial conditions, and the relationship to edge detection using filters. It is straightforward to generalize the algorithm to multiband segmentation and we demonstrate it on gray level images, color images and texture images. The novel color model allows us to eliminate intensity gradients and shadows, thereby obtaining segmentation based on the albedos of objects. It also helps detect highlight regions.