A Survey on Segmentation Techniques for Image Processing

The segmentation methods for image processing are studied in the presented work. Image segmentation can be defined as a vital step in digital image processing. Also, it is used in various applications including object co-segmentation, recognition tasks, medical imaging, content based image retrieval, object detection, machine vision and video surveillance. A lot of approaches were created for image segmentation. In addition, the main goal of segmentation is to facilitate and alter the image representation into something which is more important and simply to be analyzed. The approaches of image segmentation are splitting the images into a few parts on the basis of image’s features including texture, color, pixel intensity value and so on. With regard to the presented study, many approaches of image segmentation are reviewed and discussed. The techniques of segmentation might be categorized into six classes: First, thresholding segmentation techniques such as global thresholding (iterative thresholding, minimum error thresholding, otsu's, optimal thresholding, histogram concave analysis and entropy based thresholding), local thresholding (Sauvola’s approach, T.R Singh’s approach, Niblack’s approaches, Bernsen’s approach Bruckstein’s and Yanowitz method and Local Adaptive Automatic Binarization) and dynamic thresholding. Second, edge-based segmentation techniques such as gray-histogram technique, gradient based approach (laplacian of gaussian, differential coefficient approach, canny approach, prewitt approach, Roberts approach and sobel approach). Thirdly, region based segmentation approaches including Region growing techniques (seeded region growing (SRG), statistical region growing, unseeded region growing (UsRG)), also merging and region splitting approaches. Fourthly, clustering approaches, including soft clustering (fuzzy C-means clustering (FCM)) and hard clustering (K-means clustering). Fifth, deep neural network techniques such as convolution neural network, recurrent neural networks (RNNs), encoder-decoder and Auto encoder models and support vector machine. Finally, hybrid techniques such as evolutionary approaches, fuzzy logic and swarm intelligent (PSO and ABC techniques) and discusses the pros and cons of each method.

Automated tumor segmentation in MR brain image using fuzzy c-means clustering and seeded region methodology

<span id="docs-internal-guid-c8cba487-7fff-2314-f38a-f2936a74e0fd"><span>Automated segmentation of a tumor is still a considerably exciting research topic in the medical imaging processing field, and it plays a considerable role in forming a right diagnosis, to aid effective medical treatment. In this work, a fully automated system for segmentation of the brain tumor in MRI images is introduced. The suggested system consists of three parts: Initially, the image is pre-processed to enhance contrast, eliminate noise, and strip the skull from the image using filtering and morphological operations. Secondly, segmentation of the image happens using two techniques, fuzzy c-means clustering (FCM) and with the application of a seeded region growing algorithm (SGR). Thirdly, this method proposes a post-processing step to smooth segmentation region edges using morphological operations. The testing of the proposed system involved 233 patients, which included 287 MRI images. A comparison of the results ensued, with the manual verification of the traces performed by doctors, which ultimately proved an average Dice Coefficient of 90.13% and an average Jaccard Coefficient of 82.60% also, by comparison with traditional segmentation techniques such as FCM method. The segmentation results and quantitative data analysis demonstrates the effectiveness of the suggested system.</span></span>

Skin Lesion Segmentation Techniques for Melanoma Diagnosis: Comparative Studies

In recent years, melanoma skin cancer has been one of the rapidest risings of all cancers, which has a high risk of spread. This deadliest form of skin cancer must be diagnosed early for effective treatment. Due to the difficulty and subjectivity of human interpretation, computerized analysis of dermoscopy images has become an important research area. One of the most important steps in dermoscopy image analysis is the segmentation of skin lesion. In this research, we create different new algorithms for the skin lesion segmentation in dermoscopic images. The segmentation algorithms compared are a modified automatic Seeded Region Growing based on Particle swarm optimization image clustering which was used for breast MRI Tumours segmentation, Generalized rough fuzzy c-means algorithm which has been used for brain MR image segmentation, a Support Vector Machine (SVM) and Self-Organizing Map (SOM) with Genetic Algorithm. We used two different datasets with their masks to evaluate the accuracy, sensitivity, and specificity of various segmentation techniques. The results shows that a modified automatic Seeded Region Growing based on Particle swarm optimization image clustering has the highest accuracy (92%) compares with the other algorithms.

Skin Lesion Segmentation Techniques for Melanoma Diagnosis: Comparative Studies

In recent years, melanoma skin cancer has been one of the rapidest risings of all cancers, which has a high risk of spread. This deadliest form of skin cancer must be diagnosed early for effective treatment. Due to the difficulty and subjectivity of human interpretation, computerized analysis of dermoscopy images has become an important research area. One of the most important steps in dermoscopy image analysis is the segmentation of skin lesion. In this research, we create different new algorithms for the skin lesion segmentation in dermoscopic images. The segmentation algorithms compared are a modified automatic Seeded Region Growing based on Particle swarm optimization image clustering which was used for breast MRI Tumours segmentation, Generalized rough fuzzy c-means algorithm which has been used for brain MR image segmentation, a Support Vector Machine (SVM) and Self-Organizing Map (SOM) with Genetic Algorithm. We used two different datasets with their masks to evaluate the accuracy, sensitivity, and specificity of various segmentation techniques. The results shows that a modified automatic Seeded Region Growing based on Particle swarm optimization image clustering has the highest accuracy (92%) compares with the other algorithms.

Using multispectral airborne LiDAR data for land/water discrimination: a case study at Lake Ontario, Canada

Coastal areas are environmentally sensitive and are affected by nature events and human activities. Land/water interaction in coastal areas changes over time and, therefore, requires accurate detection and frequent monitoring. Multispectral Light Detection and Ranging (LiDAR) systems, which operate at different wavelengths, have become available. This new technology can provide an effective and accurate solution for the determination of the land/water interface. In this context, we aim to investigate a set of point features based on elevation, intensity, and geometry for this application, followed by a presentation of an unsupervised land/water discrimination method based on seeded region growing algorithm. The multispectral airborne LiDAR sensor, the Optech Titan, was used to acquire LiDAR data at three wavelengths (1550, 1064, and 532 nm) of a study area covering part of Lake Ontario in Scarborough, Canada for testing the discrimination methods. The elevation- and geometry-based features achieved an average overall accuracy of 75.1% and 74.2%, respectively, while the intensity-based features achieved 63.9% accuracy. The region growing method succeeded in discriminating water from land with more than 99% overall accuracy, and the land/water boundary was delineated with an average root mean square error of 0.51 m. The automation of this method is restricted by having double returns from water bodies at the 532 nm wavelength.

Using multispectral airborne LiDAR data for land/water discrimination: a case study at Lake Ontario, Canada

Coastal areas are environmentally sensitive and are affected by nature events and human activities. Land/water interaction in coastal areas changes over time and, therefore, requires accurate detection and frequent monitoring. Multispectral Light Detection and Ranging (LiDAR) systems, which operate at different wavelengths, have become available. This new technology can provide an effective and accurate solution for the determination of the land/water interface. In this context, we aim to investigate a set of point features based on elevation, intensity, and geometry for this application, followed by a presentation of an unsupervised land/water discrimination method based on seeded region growing algorithm. The multispectral airborne LiDAR sensor, the Optech Titan, was used to acquire LiDAR data at three wavelengths (1550, 1064, and 532 nm) of a study area covering part of Lake Ontario in Scarborough, Canada for testing the discrimination methods. The elevation- and geometry-based features achieved an average overall accuracy of 75.1% and 74.2%, respectively, while the intensity-based features achieved 63.9% accuracy. The region growing method succeeded in discriminating water from land with more than 99% overall accuracy, and the land/water boundary was delineated with an average root mean square error of 0.51 m. The automation of this method is restricted by having double returns from water bodies at the 532 nm wavelength.

WTS: A Weakly towards Strongly Supervised Learning Framework for Remote Sensing Land Cover Classification Using Segmentation Models

Land cover classification is one of the most fundamental tasks in the field of remote sensing. In recent years, fully supervised fully convolutional network (FCN)-based semantic segmentation models have achieved state-of-the-art performance in the semantic segmentation task. However, creating pixel-level annotations is prohibitively expensive and laborious, especially when dealing with remote sensing images. Weakly supervised learning methods from weakly labeled annotations can overcome this difficulty to some extent and achieve impressive segmentation results, but results are limited in accuracy. Inspired by point supervision and the traditional segmentation method of seeded region growing (SRG) algorithm, a weakly towards strongly (WTS) supervised learning framework is proposed in this study for remote sensing land cover classification to handle the absence of well-labeled and abundant pixel-level annotations when using segmentation models. In this framework, only several points with true class labels are required as the training set, which are much less expensive to acquire compared with pixel-level annotations through field survey or visual interpretation using high-resolution images. Firstly, they are used to train a Support Vector Machine (SVM) classifier. Once fully trained, the SVM is used to generate the initial seeded pixel-level training set, in which only the pixels with high confidence are assigned with class labels whereas others are unlabeled. They are used to weakly train the segmentation model. Then, the seeded region growing module and fully connected Conditional Random Fields (CRFs) are used to iteratively update the seeded pixel-level training set for progressively increasing pixel-level supervision of the segmentation model. Sentinel-2 remote sensing images are used to validate the proposed framework, and SVM is selected for comparison. In addition, FROM-GLC10 global land cover map is used as training reference to directly train the segmentation model. Experimental results show that the proposed framework outperforms other methods and can be highly recommended for land cover classification tasks when the pixel-level labeled datasets are insufficient by using segmentation models.

A Clustering Based Image Segmentation Procedure to Automatically Detect Grains in Polycrystalline Materials

The physical and mechanical properties of a polycrystalline material depend on its microstructure characteristics such as the size and morphology of grains. In practice, different imaging methods are used to visualize the grain structure of such materials. To analyze microstructural changes in case of applied stress and to predict its performance in a given application, the quantitative information about the grain structure must be taken into account. In this work, an effcient and reproducible algorithm, which automatically detects grains in different types of microstructure images, is proposed. Due to the diversity between the analyzed images and a limited number of labeled data, a clustering patch-based approach is followed. The algorithm aims to distinguish between patches in homogeneous grain areas and those lying on grain boundaries through Gaussian Mixture Modeling. The identified groups of grain patches are used to create the seed image for a Seeded Region Growing algorithm, enabling nally a pixelwise image segmentation.