Search

Light Detection and Ranging (LiDAR) sensors emit laser signals to calculate distances based on the time delay of the returned laser pulses. They can generate dense point clouds to map forest structures at a high level of spatial resolution. In this work, we consider the problem of segmenting out individual trees in Airborne Laser Scanning (ALS) point clouds. Several techniques have been proposed for this purpose which generally require time-consuming parameter tuning and intense user interaction. Our goal is to design an automated, intuitive, and robust approach requiring minimal user interaction. To this aim, we define a new segmentation approach based on topological tools, namely on the watershed transform and on persistence-based simplification. The approach follows a divide-and-conquer paradigm, splitting a LiDAR point cloud into regions with uniform densities. Our algorithm is validated on coniferous forests collected in the NEW technologies for a better mountain FORest timber mobilization (NEWFOR) dataset, and deciduous forests collected in the Smithsonian Environmental Research Center (SERC) dataset. When compared to four state-of-the-art tree segmentation algorithms, our method performs best in both ecosystem types. It provides more accurate stem estimations and single tree segmentation results at various of stem and point densities. Also, our method requires only a single (Boolean) parameter, which makes it extremely easy to use and very promising for various forest analysis applications, such as biomass estimation and field inventory surveys. [ABSTRACT FROM AUTHOR]

The largest organ in the body is the liver and primarily helps in metabolism and detoxification. Liver segmentation is a crucial step in liver cancer detection in computer vision-based biomedical image analysis. Liver segmentation is a critical task and results in under-segmentation and over-segmentation due to the complex structure of abdominal computed tomography (CT) images, noise, and textural variations over the image. This paper presents liver segmentation in abdominal CT images using marker-based watershed transforms. In the pre-processing stage, a modified double stage gaussian filter (MDSGF) is used to enhance the contrast, and preserve the edge and texture information of liver CT images. Further, marker controlled watershed transform is utilized for the segmentation of liver images from the abdominal CT images. Liver segmentation using suggested MDSGF and marker-based watershed transform help to diminish the under-segmentation and over-segmentation of the liver object. The performance of the proposed system is evaluated on the LiTS dataset based on Dice score (DS), relative volume difference (RVD), volumetric overlapping error (VOE), and Jaccard index (JI). The proposed method gives (Dice score of 0.959, RVD of 0.09, VOE of 0.089, and JI of 0.921). [ABSTRACT FROM AUTHOR]

Cereal grains are a vital part of the human diet. The appearance quality and size distribution of cereal grains play major roles as deciders or indicators of market acceptability, storage stability, and breeding. Computer vision is popular in completing quality assessment and size analysis tasks, in which an accurate instance segmentation is a key step to guaranteeing the smooth completion of tasks. This study proposes a fast deployable instance segmentation method based on a generative marker-based watershed segmentation algorithm, which combines two strategies (one strategy for optimizing kernel areas and another for comprehensive segmentation) to overcome the problems of over-segmentation and under-segmentation for images with dense and small targets. Results show that the average segmentation accuracy of our method reaches 98.73%, which is significantly higher than the marker-based watershed segmentation algorithm (82.98%). To further verify the engineering practicality of our method, we count the size distribution of segmented cereal grains. The results keep a high degree of consistency with the manually sketched ground truth. Moreover, our proposed algorithm framework can be used as a great reference in other segmentation tasks of dense targets. [ABSTRACT FROM AUTHOR]

Background: Skin micro‐relief has been researched by a variety of devices and methods, which usually are expensive or complicated. On the other hand, skin micro‐relief relates to quite a few parameters, and it is hard to evaluate all of them at the same time. In the study, all parameters related to skin micro‐relief are extracted and evaluated by image analysis. Materials and Methods: Skin micro‐relief evaluation was divided into four aspects: (a) Tamura features method was used to evaluate skin surface. (b) Morphological transform was applied to extract skin pores. (c) Watershed transform was applied to extract skin furrows. (d) labeling operation was used to evaluate the number, area and average area of skin closed polygons. Then, cheek images from 163 healthy Japanese females (0‐70 years old) are analyzed to explore the age‐dependent changes. Results: Most parameters increased as age went on with significant differences, such as skin surface coarseness, contrast, skin pore number, area, average area, skin furrow width, skin closed polygon area and skin closed polygon average area. Skin coarseness has a strong correlation with pore area. Conclusion: The method proposed in the study provided a comprehensive and effective assessment of skin micro‐relief. [ABSTRACT FROM AUTHOR]

The high computational cost of the superpixel segmentation algorithms for hyperspectral remote sensing images makes them ideal candidates for parallel computation. The waterpixel algorithm, in particular, extracts segmentation regions called waterpixels and consists of four stages called vectorial gradient, spatial regularization, marker selection, and watershed transform. In this paper, an efficient version of a GPU algorithm for waterpixel segmentation using the Compute Unified Device Architecture (CUDA) is presented. The algorithm extracts all the spectral information available in the bands of the hyperspectral image through the vectorial gradient. A cellular automaton is selected for the computation of the watershed transform using a block-asynchronous implementation with 8-connectivity. The experimental analysis shows high speedup values for the resulting GPU algorithm when it is compared to a multicore OpenMP implementation using 8 threads. [ABSTRACT FROM AUTHOR]

Unexploded ordnance (UXO) survey is the foremost task of clearance project. Transient electromagnetic method (TEM) is proved effective for UXO survey. However, it is still difficult for TEM to detect small-size, ultra-shallow and dense UXO targets because the large-size devices and inversion methods for large-scale applications are usually ineffective. In the work, in order to avoid the complex inversion of apparent resistivity, the voltages acquired by our specified small-loop TEM system are used to depict a voltage distribution profile, which is then processed with an improved imaging algorithm. Several major influences on the secondary voltage image are discussed during UXO survey, including emitting coil size, target attitude, basin effect and shell thickness. Furthermore, an improved watershed imaging algorithm is proposed to obtain the best threshold marking and automatically realize the dense target discrimination. Over-segmentation resulting from burrs, noises, and scraps is reduced through background marking. And also under-segmentation is avoided during foreground marking through adaptive search of saddle point. The algorithm has been verified effective using simulation data and experiment data of UXO-like targets.

Cereal grains are a vital part of the human diet. The appearance quality and size distribution of cereal grains play major roles as deciders or indicators of market acceptability, storage stability, and breeding. Computer vision is popular in completing quality assessment and size analysis tasks, in which an accurate instance segmentation is a key step to guaranteeing the smooth completion of tasks. This study proposes a fast deployable instance segmentation method based on a generative marker-based watershed segmentation algorithm, which combines two strategies (one strategy for optimizing kernel areas and another for comprehensive segmentation) to overcome the problems of over-segmentation and under-segmentation for images with dense and small targets. Results show that the average segmentation accuracy of our method reaches 98.73%, which is significantly higher than the marker-based watershed segmentation algorithm (82.98%). To further verify the engineering practicality of our method, we count the size distribution of segmented cereal grains. The results keep a high degree of consistency with the manually sketched ground truth. Moreover, our proposed algorithm framework can be used as a great reference in other segmentation tasks of dense targets.

Liver cancer is the main source of death in the globe. Manual cancer tissue diagnosis is monotonous and troublesome. Hence, the paper fosters a high-exactness automatic diagnosis strategy for liver cancer growth. The image processing approach can utilize Computer Aided Diagnosis (CAD) for the arrangement of liver malignant growth to help the specialist. The CAD system is used to give a robotized approach to deal with successful arrangement of liver malignancy using feasible arrangements. Early affirmation and finding of liver growth are crucial for the space of liver cancers. Medical image processing is utilized to isolate tumors in a non-prominent way. Different strategies for recognizing liver tumors dependent upon abnormal lesion size and shape have been made. In like manner, automatic procedures for dividing the liver and liver tumors are pursued in clinical practice. This paper examines out a combination of liver malignant growth determination algorithms and philosophies. [ABSTRACT FROM AUTHOR]

Background and Aim: deep learning has not been successfully implemented in liver tumour feature extraction and classification using computer-aided diagnosis. This study aims to enhance classification accuracy and improves the processing time to better differentiate tumour types. Methodology: This study proposed a hybrid model, which combines the regularization function with the current loss function for the support vector machine (SVM) classifier. Regularization function is used for prioritizing image classes before feeding it to the linear mapping. The proposed model consists of the region growing algorithm to get the region-of-interest (ROI), and Weiner filtering algorithm for image enhancement and noise removal. The gray level co-occurrence matrix (GLCM) was performed to extract the feature from the image. The extracted feature then fed to SVM classifier using selected feature vectors to classify the affected region and neglecting the unwanted areas. Results: classification accuracy was calculated using probability score, and the processing time was calculated based on the total execution time. The proposed system was able to achieve an average classification accuracy of 98.9%, which is about 2–3% higher than the current system. The results showed that 12 ms reduced the processing time on average. Conclusion: The proposed system focused on improving feature extraction and classification for different types of tumours from the MRI images. The study solved the problem in linear mapping of support vector machine and enhanced the classification accuracy and the processing time of early diagnosis of three different types of tumours in liver MRI images. [ABSTRACT FROM AUTHOR]

• We introduce the use of hierarchical Watershed Transform for point cloud segmentation. • Our approach combines color and geometric information to segment architectural elements. • We provide a systematic review of the literature on segmentation for construction point clouds. • We propose a new technique in order to obtain the point adjacency information based on the use of an octree graph. • We create a public dataset of historical building points cloud and we perform quantitative and qualititive analysis. Segmenting accurately point clouds is of great relevance in several fields of engineering and construction. Users are interested in properly dividing an point cloud into their components and then recognizing them. Point clouds representing historical buildings present an additional challenge because image details could be related to a cultural or architectural aspect. Therefore, the way the results are evaluated is also important. In this paper, we present a novel point cloud approach for segmenting historical building of different architectural styles and periods. In our approach, that works for organized and unorganized point clouds, we combine Hierarchical Watershed Transform and curvature analysis from region growing methods in order to obtain more suitable seeds. Experiments were conducted involving historical building acquired using drones and terrestrial laser scanner. The data was combined into a single point cloud. Finally, we evaluated our results qualitatively and quantitatively, by comparing them to a dataset containing the ground truth. The quantitative metrics demonstrate the effectiveness of our method when compared with state-of-the-art methods. [ABSTRACT FROM AUTHOR]

Hyperspectral imaging is a new technology which has the ability to obtain both appearance information and spectral information. It has been successfully used in remote sensing fields such as military, agricultural and atmospheric detection. And microscopic hyperspectral imaging, as a new way for us to explore the microscopic world, has also been successfully applied in the fields of cell classification, food safety, cancer diagnosis and so on. However, the computation cost of the direct segmentation or classification for hyperspectral images in high-dimensional spectral eigenspace is usually very high. One way to decrease the cost is to reduce the hyperspectral data dimensionality via dimension-reduction algorithms before segmentation. Unfortunately, some detailed information of the hyperspectral image will be lost due to such dimension reduction, resulting in inevitably impact to the image segmentation. To solve this problem, a new hyperspectral image segmentation method is proposed in this paper. A concept of spectral angle space is established in the first, then all the pixels in the image can be projected into the space to be a sample point by calculating the spectral angles between the pixel and its adjacent pixels. The distances from the sample points to the original point in spectral angle space are calculated and mapped to gray values resulting in a grayscale image with the edge information highlighted. Subsequently the generated gray image is segmented by watershed transform and the spectral vectors of the local minima of all the segmented regions are extracted for comparison and analysis, Over-segmentation problem is solved by merging the regions together according to the similarity of the local minima pixel spectral vectors. Lastly, the proposed hyperspectral image segmentation method is experimentally validated and analyzed via simulated and actual hyperspectral data sets. [ABSTRACT FROM AUTHOR]

Cervical cancer can be prevented by having regular screenings to find any precancers and treat them. The Pap test looks for any abnormal or precancerous changes in the cells on the cervix. However, the manual screening of Pap smear in the microscope is subjective with poorly reproducible criteria. Therefore, the aim of this study was to develop a computer-assisted screening system for cervical cancer using digital image processing of Pap smear images. The analysis of Pap smear image is important in the cervical cancer screening system. There were four basic steps in our cervical cancer screening system. In cell segmentation, nuclei were detected using a shape-based iterative method, and the overlapping cytoplasm was separated using a marker-control watershed approach. In the features extraction step, three important features were extracted from the regions of segmented nuclei and cytoplasm. RF (random forest) algorithm was used as a feature selection method. In the classification stage, bagging ensemble classifier, which combined the results of five classifiers—LD (linear discriminant), SVM (support vector machine), KNN (k-nearest neighbor), boosted trees, and bagged trees—was applied. SIPaKMeD and Herlev datasets were used to prove the effectiveness of our proposed system. According to the experimental results, 98.27% accuracy in two-class classification and 94.09% accuracy in five-class classification was achieved using the SIPaKMeD dataset. When the results were compared with five classifiers, our proposed method was significantly better in two-class and five-class problems. [ABSTRACT FROM AUTHOR]

Many studies have focused on performing variational-scale segmentation to represent various geographical objects in high-resolution remote-sensing images. However, it remains a significant challenge to select the most appropriate scales based on the geographical-distribution characteristics of ground objects. In this study, we propose a variational-scale multispectral remote-sensing image segmentation method using spectral indices. Real scenes in remote-sensing images contain different types of land cover with different scales. Therefore, it is difficult to segment images optimally based on the scales of different ground objects. To guarantee image segmentation of ground objects with their own scale information, spectral indices that can be used to enhance some types of land cover, such as green cover and water bodies, were introduced into marker generation for the watershed transformation. First, a vector field model was used to determine the gradient of a multispectral remote-sensing image, and a marker was generated from the gradient. Second, appropriate spectral indices were selected, and the kernel density estimation was used to generate spectral-index marker images based on the analysis of spectral indices. Third, a series of mathematical morphology operations were used to obtain a combined marker image from the gradient and the spectral index markers. Finally, the watershed transformation was used for image segmentation. In a segmentation experiment, an optimal threshold for the spectral-index-marker generation method was identified. Additionally, the influence of the scale parameter was analyzed in a segmentation experiment based on a five-subset dataset. The comparative results for the proposed method, the commonly used watershed segmentation method, and the multiresolution segmentation method demonstrate that the proposed method yielded multispectral remote-sensing images with much better performance than the other methods.

Abstract Background Image segmentation is an essential and non trivial task in computer vision and medical image analysis. Computed tomography (CT) is one of the most accessible medical examination techniques to visualize the interior of a patient’s body. Among different computer-aided diagnostic systems, the applications dedicated to kidney segmentation represent a relatively small group. In addition, literature solutions are verified on relatively small databases. The goal of this research is to develop a novel algorithm for fully automated kidney segmentation. This approach is designed for large database analysis including both physiological and pathological cases. Methods This study presents a 3D marker-controlled watershed transform developed and employed for fully automated CT kidney segmentation. The original and the most complex step in the current proposition is an automatic generation of 3D marker images. The final kidney segmentation step is an analysis of the labelled image obtained from marker-controlled watershed transform. It consists of morphological operations and shape analysis. The implementation is conducted in a MATLAB environment, Version 2017a, using i.a. Image Processing Toolbox. 170 clinical CT abdominal studies have been subjected to the analysis. The dataset includes normal as well as various pathological cases (agenesis, renal cysts, tumors, renal cell carcinoma, kidney cirrhosis, partial or radical nephrectomy, hematoma and nephrolithiasis). Manual and semi-automated delineations have been used as a gold standard. Wieclawek Among 67 delineated medical cases, 62 cases are ‘Very good’, whereas only 5 are ‘Good’ according to Cohen’s Kappa interpretation. The segmentation results show that mean values of Sensitivity, Specificity, Dice, Jaccard, Cohen’s Kappa and Accuracy are 90.29, 99.96, 91.68, 85.04, 91.62 and 99.89% respectively. All 170 medical cases (with and without outlines) have been classified by three independent medical experts as ‘Very good’ in 143–148 cases, as ‘Good’ in 15–21 cases and as ‘Moderate’ in 6–8 cases. Conclusions An automatic kidney segmentation approach for CT studies to compete with commonly known solutions was developed. The algorithm gives promising results, that were confirmed during validation procedure done on a relatively large database, including 170 CTs with both physiological and pathological cases.

This paper proposes a novel approach for living and missing vine identification and vine characterization in goblet-trained vine plots using aerial images. Given the periodic structure of goblet vineyards, the RGB color coded parcel image is analyzed using proper processing techniques in order to determine the locations of living and missing vines. Vine characterization is achieved by implementing the marker-controlled watershed transform where the centers of the living vines serve as object markers. As a result, a precise mortality rate is calculated for each parcel. Moreover, all vines, even the overlapping ones, are fully recognized providing information about their size, shape, and green color intensity. The presented approach is fully automated and yields accuracy values exceeding 95% when the obtained results are assessed with ground-truth data. This unsupervised and automated approach can be applied to any type of plots presenting similar spatial patterns requiring only the image as input.

Abstract Objectives Liver cancer is one of the leading cause of death in all over the world. Detecting the cancer tissue manually is a difficult task and time consuming. Hence, a computer-aided diagnosis (CAD) is used in decision making process for accurate detection for appropriate therapy. Therefore the main objective of this work is to detect the liver cancer accurately using automated method. Methods In this work, we have proposed a new system called as watershed Gaussian based deep learning (WGDL) technique for effective delineate the cancer lesion in computed tomography (CT) images of the liver. A total of 225 images were used in this work to develop the proposed model. Initially, the liver was separated using marker controlled watershed segmentation process and finally the cancer affected lesion was segmented using the Gaussian mixture model (GMM) algorithm. After tumor segmentation, various texture features were extracted from the segmented region. These segmented features were fed to deep neural network (DNN) classifier for automated classification of three types of liver cancer i.e. hemangioma (HEM), hepatocellular carcinoma (HCC) and metastatic carcinoma (MET). Results We have achieved a classification accuracy of 99.38%, Jaccard index of 98.18%, at 200 epochs using DNN classifier with a negligible validation loss of 0.062 during the classification process. Conclusions Our developed system is ready to be tested with huge database and can aid the radiologist in detecting the liver cancer using CT images. [ABSTRACT FROM AUTHOR]

The traditional watershed algorithm has the limitation of false mark in medical image segmentation, which causes over-segmentation and images to be contaminated by noise possibly during acquisition. In this study, we proposed an improved watershed segmentation algorithm based on morphological processing and total variation model (TV) for medical image segmentation. First of all, morphological gradient preprocessing is performed on MRI images of brain lesions. Secondly, the gradient images are denoised by the all-variational model. While retaining the edge information of MRI images of brain lesions, the image noise is reduced. And then, the internal and external markers are obtained by forced minimum technique, and the gradient amplitude images are corrected by using these markers. Finally, the modified gradient image is subjected to watershed transformation. The experiment of segmentation and simulation of brain lesion MRI image is carried out on MATLAB. And the segmentation results are compared with other watershed algrothims. The experimental results demonstrate that our method obtains the least number of regions, which can extract MRI images of brain lesions effectively. In addition, this method can inhibit over-segmentation, improving the segmentation results of lesions in MRI images of brain lesions. [ABSTRACT FROM AUTHOR]

In this paper we introduce a novel machine learning-based fully automatic approach for the semantic analysis and documentation of masonry wall images, performing in parallel automatic detection and virtual completion of occluded or damaged wall regions, and brick segmentation leading to an accurate model of the wall structure. For this purpose, we propose a four-stage algorithm which comprises three interacting deep neural networks and a watershed transform-based brick outline extraction step. At the beginning, a U-Net-based sub-network performs initial wall segmentation into brick, mortar and occluded regions, which is followed by a two-stage adversarial inpainting model. The first adversarial network predicts the schematic mortar-brick pattern of the occluded areas based on the observed wall structure, providing in itself valuable structural information for archeological and architectural applications. The second adversarial network predicts the pixels’ color values yielding a realistic visual experience for the observer. Finally, using the neural network outputs as markers in a watershed-based segmentation process, we generate the accurate contours of the individual bricks, both in the originally visible and in the artificially inpainted wall regions. Note that while the first three stages implement a sequential pipeline, they interact through dependencies of their loss functions admitting the consideration of hidden feature dependencies between the different network components. For training and testing the network a new dataset has been created, and an extensive qualitative and quantitative evaluation versus the state-of-the-art is given. The experiments confirmed that the proposed method outperforms the reference techniques both in terms of wall structure estimation and regarding the visual quality of the inpainting step, moreover it can be robustly used for various different masonry wall types.

Motivated by applications in topographic map information extraction, our goal was to discover a practical method for scanned topographic map (STM) segmentation. We present an advanced guided watershed transform (AGWT) to generate superpixels on STM. AGWT utilizes the information from both linear and area elements to modify detected boundary maps and sequentially achieve superpixels based on the watershed transform. With achieving an average of 0.06 on under-segmentation error, 0.96 on boundary recall, and 0.95 on boundary precision, it has been proven to have strong ability in boundary adherence, with fewer over-segmentation issues. Based on AGWT, a benchmark for STM segmentation based on superpixels and a shallow convolutional neural network (SCNN), termed SSCNN, is proposed. There are several notable ideas behind the proposed approach. Superpixels are employed to overcome the false color and color aliasing problems that exist in STMs. The unification method of random selection facilitates sufficient training data with little manual labeling while keeping the potential color information of each geographic element. Moreover, with the small number of parameters, SCNN can accurately and efficiently classify those unified pixel sequences. The experiments show that SSCNN achieves an overall F1 score of 0.73 on our STM testing dataset. They also show the quality of the segmentation results and the short run time of this approach, which makes it applicable to full-size maps.

Cervical cancer can be prevented by having regular screenings to find any precancers and treat them. The Pap test looks for any abnormal or precancerous changes in the cells on the cervix. However, the manual screening of Pap smear in the microscope is subjective with poorly reproducible criteria. Therefore, the aim of this study was to develop a computer-assisted screening system for cervical cancer using digital image processing of Pap smear images. The analysis of Pap smear image is important in the cervical cancer screening system. There were four basic steps in our cervical cancer screening system. In cell segmentation, nuclei were detected using a shape-based iterative method, and the overlapping cytoplasm was separated using a marker-control watershed approach. In the features extraction step, three important features were extracted from the regions of segmented nuclei and cytoplasm. RF (random forest) algorithm was used as a feature selection method. In the classification stage, bagging ensemble classifier, which combined the results of five classifiers—LD (linear discriminant), SVM (support vector machine), KNN (k-nearest neighbor), boosted trees, and bagged trees—was applied. SIPaKMeD and Herlev datasets were used to prove the effectiveness of our proposed system. According to the experimental results, 98.27% accuracy in two-class classification and 94.09% accuracy in five-class classification was achieved using the SIPaKMeD dataset. When the results were compared with five classifiers, our proposed method was significantly better in two-class and five-class problems.

The satellite image processing is an important tool for decision making in domains like agriculture, forestry, hydrology, for normal activity tracking but also in special situations caused by natural disasters. In this paper it is proposed a method for forestry surface evaluation in terms of occupied surface and also as number of trees. The segmentation method is based on watershed transform which offers good performances in case the objects to detect have connected borders. The method is applied for automatic multi-temporal analysis of forestry areas and represents a useful instrument for decision makers.

This paper proposes an object-based segmentation/classification scheme for remotely sensed images, based on a novel variant of the recently proposed Genetic Sequential Image Segmentation (GeneSIS) algorithm. GeneSIS segments the image in an iterative manner, whereby at each iteration a single object is extracted via a genetic-based object extraction algorithm. Contrary to the previous pixel-based GeneSIS where the candidate objects to be extracted were evaluated through the fuzzy content of their included pixels, in the newly developed region-based GeneSIS algorithm, a watershed-driven fine segmentation map is initially obtained from the original image, which serves as the basis for the forthcoming GeneSIS segmentation. Furthermore, in order to enhance the spatial search capabilities, we introduce a more descriptive encoding scheme in the object extraction algorithm, where the structural search modules are represented by polygonal shapes. Our objectives in the new framework are posed as follows: enhance the flexibility of the algorithm in extracting more flexible object shapes, assure high level classification accuracies, and reduce the execution time of the segmentation, while at the same time preserving all the inherent attributes of the GeneSIS approach. Finally, exploiting the inherent attribute of GeneSIS to produce multiple segmentations, we also propose two segmentation fusion schemes that operate on the ensemble of segmentations generated by GeneSIS. Our approaches are tested on an urban and two agricultural images. The results show that region-based GeneSIS has considerably lower computational demands compared to the pixel-based one. Furthermore, the suggested methods achieve higher classification accuracies and good segmentation maps compared to a series of existing algorithms.

Individual tree crown detection and delineation (ITCD) mainly depend on high-resolution aerial photos and satellite images or LiDAR, and these data can be costly to obtain. The advent of unmanned aerial vehicle (UAV) remote sensing technology provides an economic and effective method for data acquisition. Therefore, the research of ITCD based on UAV high-resolution images is of significance to improve the efficiency and accuracy of forest resource inventory and remote sensing validation. However, in very high-resolution (defined here with the pixel size smaller than 10 cm) images, the inhomogeneities in the canopy texture can be detrimental to the correct detection of individual trees by computer processing. We applied the bias field estimation, which is used in medical image segmentation, to reduce the within-canopy spectral heterogeneity in the very high-resolution UAV-derived orthophoto. By selecting young Osmanthus and Podocarpus trees that grow in a nursery as the study objects, we tested our method in an orthophoto (with a ground resolution of 2.5 cm) generated from overlapping UAV images. A local intensity clustering was first applied to produce a smoothed bias field image; then, by using morphological operation of opening and closing, the fine texture of the canopy was further smoothed. Finally, individual tree crowns were extracted by applying the marker-controlled watershed segmentation algorithm. The segmentation results were validated by comparing to the manually drawn individual tree crown polygons. The $F$ -scores of the detection rate for Osmanthus and Podocarpus were 98.2% and 93.1%, respectively. The result is considerably better than those achieved with similar processing steps but without the bias field treatment. Our study proves that it is feasible and effective to detect and delineate individual tree crowns based on the bias field and marker-controlled watershed segmentation in very-high-resolution images obtained from UAV. [ABSTRACT FROM AUTHOR]

This paper presents a novel method to segment bone fragments imaged using 3D Computed Tomography (CT). Existing image segmentation solutions often lack accuracy when segmenting internal trabecular and cancellous bone tissues from adjacent soft tissues having similar appearance and often merge regions associated with distinct fragments. These issues create problems in downstream visualization and pre-operative planning applications and impede the development of advanced image-based analysis methods such as virtual fracture reconstruction. The proposed segmentation algorithm uses a probability-based variation of the watershed transform, referred to as the Probabilistic Watershed Transform (PWT). The PWT uses a set of probability distributions, one for each bone fragment, that model the likelihood that a given pixel is a measurement from one of the bone fragments. The likelihood distributions proposed improve upon known shortcomings in competing segmentation methods for bone fragments within CT images. A quantitative evaluation of the bone segmentation results is provided that compare our segmentation results with several leading competing methods as well as human-generated segmentations. [ABSTRACT FROM AUTHOR]

Porosity is the most crucial parameter to assess the potential of fine grain rocks like shale. Shale has very low connected porosity and permeability, which controls the fluid flow and migration. Therefore, detection of connected and closed pore network in digital images can help to evaluate porosity and permeability of rock. The objective of this paper is to characterize Gondwana shale of eastern India in micro-scale, that is, in terms of its porosity, pore-structure and pore size distribution for its shale gas potential. Watershed-transform has been performed to segment pores, throat and mineral grains by using 2-D Scan Electron Microscopy (SEM) images of Gondwana shale of Barren-Measures Formation. The watershed transform, when coupled with morphological operators and a customized disk shape structuring element, segments the pores and throats efficiently and estimates the porosity of Gondwana shale for the very first time. Three Gondwana shale samples from different borehole depth is used for this study. Finally, porosity measured in a laboratory using Mercury Injection Capillary Pressure (MICP) test, is compared with the numerically modeled watershed porosity values, which shows good agreement between the two sets of results with precisely segmented pore bodies and throat bodies at appropriate locations. Most of the pores in the sample are found to be in a mesopore category (2–50 nm). The porosity values measured using MICP ranges from 5.01% to 6.53% whereas the porosity estimated from watershed image segmentation ranges from 5.21% to 6.91%. [ABSTRACT FROM AUTHOR]

Particle shape was one of the most important factors which affects the gravel aggregate's properties. It was also one of the important factors that directly affects the performance of asphalt pavements. In this paper, the gravel aggregate of quartzite was studied by using the industrial CT instrument. MATLAB was used to capture the aggregate slice properties including reverse color, median filtering, noise reduction, binarization and so on. The 3D aggregate model was reconstructed by using the software of MIMICS. The three-dimensional model of the aggregate was further optimized. The best fitting cuboid, cylinder, cone and sphere information of the aggregate were obtained by using the characteristics analysis function. [ABSTRACT FROM AUTHOR]