Images as occlusions of textures: a framework for segmentation

Automated Layer Identification Method for Skin Tissue Histology Images

We present a novel automated tissue layer identification method for histology images. The method requires a single user input: the number of layers to be identified. The method incorporates a coarse boundary identification step followed by a refinement step. The coarse identification segments the image into 125 × 125 pixel sub-tiles, computes the histogram of each sub-tile, implements K-means clustering to label each sub-tile, and uses Dijkstra's algorithm to form the layer boundary. The refinement step identifies hair follicles, improves the detail and accuracy of the boundary, and segments the epidermis. The method only uses one color channel (blue). We test our proposed method using eight excised porcine tissue samples taken at different anatomical locations. The layer segmentations demonstrated that the dermis thickness increased, and the subcutaneous thickness decreased moving from breast to belly. Minimal variation in the thickness of the epidermis layer across anatomical locations was observed. Overall, these results highlight the importance of quantifying and assessing the tissue environment. Moreover, we demonstrate that our proposed method was robust across different histology stains and did not depend on color-specific information.

Radiomics and Digital Image Texture Analysis in Oncology (Review)

One of the most promising areas of diagnosis and prognosis of diseases is radiomics, a science combining radiology, mathematical modeling, and deep machine learning. The main concept of radiomics is image biomarkers (IBMs), the parameters characterizing various pathological changes and calculated based on the analysis of digital image texture. IBMs are used for quantitative assessment of digital imaging results (CT, MRI, ultrasound, PET). The use of IBMs in the form of “virtual biopsy” is of particular relevance in oncology.

The article provides the basic concepts of radiomics identifying the main stages of obtaining IBMs: data collection and preprocessing, tumor segmentation, data detection and extraction, modeling, statistical processing, and data validation. The authors have analyzed the possibilities of using IBMs in oncology, describing the currently known features and advantages of using radiomics and image texture analysis in the diagnosis and prognosis of cancer. The limitations and problems associated with the use of radiomics data are considered.

Although the novel effective tool for performing virtual biopsy of human tissue is at the development stage, quite a few projects have already been implemented, and medical software packages for radiomics analysis of digital images have been created.

A Class-Independent Texture-Separation Method Based on a Pixel-Wise Binary Classification

Texture segmentation is a challenging problem in computer vision due to the subjective nature of textures, the variability in which they occur in images, their dependence on scale and illumination variation, and the lack of a precise definition in the literature. This paper proposes a method to segment textures through a binary pixel-wise classification, thereby without the need for a predefined number of textures classes. Using a convolutional neural network, with an encoder–decoder architecture, each pixel is classified as being inside an internal texture region or in a border between two different textures. The network is trained using the Prague Texture Segmentation Datagenerator and Benchmark and tested using the same dataset, besides the Brodatz textures dataset, and the Describable Texture Dataset. The method is also evaluated on the separation of regions in images from different applications, namely remote sensing images and H&E-stained tissue images. It is shown that the method has a good performance on different test sets, can precisely identify borders between texture regions and does not suffer from over-segmentation.

2D Image-To-3D Model: Knowledge-Based 3D Building Reconstruction (3DBR) Using Single Aerial Images and Convolutional Neural Networks (CNNs)

In this study, a deep learning (DL)-based approach is proposed for the detection and reconstruction of buildings from a single aerial image. The pre-required knowledge to reconstruct the 3D shapes of buildings, including the height data as well as the linear elements of individual roofs, is derived from the RGB image using an optimized multi-scale convolutional–deconvolutional network (MSCDN). The proposed network is composed of two feature extraction levels to first predict the coarse features, and then automatically refine them. The predicted features include the normalized digital surface models (nDSMs) and linear elements of roofs in three classes of eave, ridge, and hip lines. Then, the prismatic models of buildings are generated by analyzing the eave lines. The parametric models of individual roofs are also reconstructed using the predicted ridge and hip lines. The experiments show that, even in the presence of noises in height values, the proposed method performs well on 3D reconstruction of buildings with different shapes and complexities. The average root mean square error (RMSE) and normalized median absolute deviation (NMAD) metrics are about 3.43 m and 1.13 m, respectively for the predicted nDSM. Moreover, the quality of the extracted linear elements is about 91.31% and 83.69% for the Potsdam and Zeebrugge test data, respectively. Unlike the state-of-the-art methods, the proposed approach does not need any additional or auxiliary data and employs a single image to reconstruct the 3D models of buildings with the competitive precision of about 1.2 m and 0.8 m for the horizontal and vertical RMSEs over the Potsdam data and about 3.9 m and 2.4 m over the Zeebrugge test data.

Object-Based Dense Matching Method for Maintaining Structure Characteristics of Linear Buildings

In this paper, we proposed a novel object-based dense matching method specially for the high-precision disparity map of building objects in urban areas, which can maintain accurate object structure characteristics. The proposed framework mainly includes three stages. Firstly, an improved edge line extraction method is proposed for the edge segments to fit closely to building outlines. Secondly, a fusion method is proposed for the outlines under the constraint of straight lines, which can maintain the building structural attribute with parallel or vertical edges, which is very useful for the dense matching method. Finally, we proposed an edge constraint and outline compensation (ECAOC) dense matching method to maintain building object structural characteristics in the disparity map. In the proposed method, the improved edge lines are used to optimize matching search scope and matching template window, and the high-precision building outlines are used to compensate the shape feature of building objects. Our method can greatly increase the matching accuracy of building objects in urban areas, especially at building edges. For the outline extraction experiments, our fusion method verifies the superiority and robustness on panchromatic images of different satellites and different resolutions. For the dense matching experiments, our ECOAC method shows great advantages for matching accuracy of building objects in urban areas compared with three other methods.

Object-Oriented Hierarchy Radiation Consistency for Different Temporal and Different Sensor Images

In the paper, we propose a novel object-oriented hierarchy radiation consistency method for dense matching of different temporal and different sensor data in the 3D reconstruction. For different temporal images, our illumination consistency method is proposed to solve both the illumination uniformity for a single image and the relative illumination normalization for image pairs. Especially in the relative illumination normalization step, singular value equalization and linear relationship of the invariant pixels is combined used for the initial global illumination normalization and the object-oriented refined illumination normalization in detail, respectively. For different sensor images, we propose the union group sparse method, which is based on improving the original group sparse model. The different sensor images are set to a similar smoothness level by the same threshold of singular value from the union group matrix. Our method comprehensively considered the influence factors on the dense matching of the different temporal and different sensor stereoscopic image pairs to simultaneously improve the illumination consistency and the smoothness consistency. The radiation consistency experimental results verify the effectiveness and superiority of the proposed method by comparing two other methods. Moreover, in the dense matching experiment of the mixed stereoscopic image pairs, our method has more advantages for objects in the urban area.

Model-based learning of local image features for unsupervised texture segmentation

Features that capture well the textural patterns of a certain class of images are crucial for the performance of texture segmentation methods. The manual selection of features or designing new ones can be a tedious task. Therefore, it is desirable to automatically adapt the features to a certain image or class of images. Typically, this requires a large set of training images with similar textures and ground truth segmentation. In this work, we propose a framework to learn features for texture segmentation when no such training data is available. The cost function for our learning process is constructed to match a commonly used segmentation model, the piecewise constant Mumford-Shah model. This means that the features are learned such that they provide an approximately piecewise constant feature image with a small jump set. Based on this idea, we develop a two-stage algorithm which first learns suitable convolutional features and then performs a segmentation. We note that the features can be learned from a small set of images, from a single image, or even from image patches. The proposed method achieves a competitive rank in the Prague texture segmentation benchmark, and it is effective for segmenting histological images.

Spatial Statistics for Segmenting Histological Structures in H&E Stained Tissue Images

Segmenting a broad class of histological structures in transmitted light and/or fluorescence-based images is a prerequisite for determining the pathological basis of cancer, elucidating spatial interactions between histological structures in tumor microenvironments (e.g., tumor infiltrating lymphocytes), facilitating precision medicine studies with deep molecular profiling, and providing an exploratory tool for pathologists. This paper focuses on segmenting histological structures in hematoxylin- and eosin-stained images of breast tissues, e.g., invasive carcinoma, carcinoma in situ, atypical and normal ducts, adipose tissue, and lymphocytes. We propose two graph-theoretic segmentation methods based on local spatial color and nuclei neighborhood statistics. For benchmarking, we curated a data set of 232 high-power field breast tissue images together with expertly annotated ground truth. To accurately model the preference for histological structures (ducts, vessels, tumor nets, adipose, etc.) over the remaining connective tissue and non-tissue areas in ground truth annotations, we propose a new region-based score for evaluating segmentation algorithms. We demonstrate the improvement of our proposed methods over the state-of-the-art algorithms in both region- and boundary-based performance measures.

A Hierarchical Building Segmentation in Digital Surface Models for 3D Reconstruction

In this study, a hierarchical method for segmenting buildings in a digital surface model (DSM), which is used in a novel framework for 3D reconstruction, is proposed. Most 3D reconstructions of buildings are model-based. However, the limitations of these methods are overreliance on completeness of the offline-constructed models of buildings, and the completeness is not easily guaranteed since in modern cities buildings can be of a variety of types. Therefore, a model-free framework using high precision DSM and texture-images buildings was introduced. There are two key problems with this framework. The first one is how to accurately extract the buildings from the DSM. Most segmentation methods are limited by either the terrain factors or the difficult choice of parameter-settings. A level-set method are employed to roughly find the building regions in the DSM, and then a recently proposed 'occlusions of random textures model' are used to enhance the local segmentation of the buildings. The second problem is how to generate the facades of buildings. Synergizing with the corresponding texture-images, we propose a roof-contour guided interpolation of building facades. The 3D reconstruction results achieved by airborne-like images and satellites are compared. Experiments show that the segmentation method has good performance, and 3D reconstruction is easily performed by our framework, and better visualization results can be obtained by airborne-like images, which can be further replaced by UAV images.

Factorization-Based Texture Segmentation

This paper introduces a factorization-based approach that efficiently segments textured images. We use local spectral histograms as features, and construct an M × N feature matrix using M-dimensional feature vectors in an N-pixel image. Based on the observation that each feature can be approximated by a linear combination of several representative features, we factor the feature matrix into two matrices--one consisting of the representative features and the other containing the weights of representative features at each pixel used for linear combination. The factorization method is based on singular value decomposition and nonnegative matrix factorization. The method uses local spectral histograms to discriminate region appearances in a computationally efficient way and at the same time accurately localizes region boundaries. The experiments conducted on public segmentation data sets show the promise of this simple yet powerful approach.

Unsupervised segmentation of low-contrast multichannel images: discrimination of tissue components in microscopic images of unstained specimens

Low-contrast images, such as color microscopic images of unstained histological specimens, are composed of objects with highly correlated spectral profiles. Such images are very hard to segment. Here, we present a method that nonlinearly maps low-contrast color image into an image with an increased number of non-physical channels and a decreased correlation between spectral profiles. The method is a proof-of-concept validated on the unsupervised segmentation of color images of unstained specimens, in which case the tissue components appear colorless when viewed under the light microscope. Specimens of human hepatocellular carcinoma, human liver with metastasis from colon and gastric cancer and mouse fatty liver were used for validation. The average correlation between the spectral profiles of the tissue components was greater than 0.9985, and the worst case correlation was greater than 0.9997. The proposed method can potentially be applied to the segmentation of low-contrast multichannel images with high spatial resolution that arise in other imaging modalities.