Watershed Method Research Articles

Pemisahan Objek Sel Tumpang Tindih pada Citra Pap Smear dengan Metode Deep learning dan Watershed

The object observed in the Pap Smear image is Cervical Cancer which forms overlapping cells. This cancer must be observed early because it is a disease that causes the death of thousands of women worldwide every year. The death rate from this disease is the fourth highest among women. One way to be aware of this disease is to do an early check on the Pap Smear test image. This cell separation process uses the image segmentation method. This method is one of the important steps to be able to identify existing cell objects. This study proposes a segmentation method to separate 2 overlapping cells in the RepomedUNM dataset. The dataset is engineered in the manufacture of synthetic Pap Smear images. The segmentation method proposed is a Deep Learning-based method so that it can identify 2 overlapping cells in one area. The level of accuracy of the test with an average score of Intersection over Union (IoU) is 0.9003. And the results of segmentation with Deep Learning can be divided into all areas using the Watershed segmentation method. So that this research becomes a reference in the early identification of Cervical Cancer.

TBC Bacteria Detection in Microscopic Image With Watershed Countur Method

Tuberculosis (TB) is an infectious disease that can be detected using a sputum sample. TB cases in Indonesia have spread throughout the region; the highest cases are in West Java. This problem makes the government do some handling and prevention of TB disease. The Bandung City Health Office (DKKB) conducted a cross-test to diagnose TB using a sputum sample. So in this study, a TB bacteria detection system, namely Mycobacterium Tuberculosis (MTB), will be made in sputum samples and their number to diagnose TB. Detection and calculation of the number of MTB are done by processing the image on the sputum sample using the watershed contour detection method. In this study, sputum sample data were obtained from DKKB. The acquisition of microscopic images at each point of the field of view is carried out using an SLR camera connected directly to the microscope to replace the function of the ocular lens. In this study, the microscopic sputum sample images were classified into positive and negative using the watershed and colorspace methods and were tested on a total of 90 microscopic images. From the system testing results, the system accuracy level is 100%, and the system precision is 100% for the detection of TB diagnosis. The system processing time averaged 5.811 seconds for 90 images used.

A Robust Approach for Segmentation and Classification of Lung Cancer using Marker Controlled Watershed Method and Deep Hybrid Learning

A Robust Approach for Segmentation and Classification of Lung Cancer using Marker Controlled Watershed Method and Deep Hybrid Learning

A new procedure for lung region segmentation from computed tomography images

<span lang="EN-US">Lung cancer is the leading cause of cancer death among people worldwide. The primary aim of this research is to establish an image processing method for lung cancer detection. This paper focuses on lung region segmentation from computed tomography (CT) scan images. In this work, a new procedure for lung region segmentation is proposed. First, the lung CT scan images will undergo an image thresholding stage before going through two morphological reconstruction and masking stages. In between morphological and masking stages, object extraction, border change, and object elimination will occur. Finally, the lung field will be annotated. The outcomes of the proposed procedure and previous lung segmentation methods i.e., the modified watershed segmentation method is compared with the ground truth images for performance evaluation that will be carried out both in qualitative and quantitative manners. Based on the analyses, the new proposed procedure for lung segmentation, denotes better performance, an increment by 0.02% to 3.5% in quantitative analysis. The proposed procedure produced better-segmented images for qualitative analysis and became the most frequently selected method by the 22 experts. This study shows that the outcome from the proposed method outperforms the existing modified watershed segmentation method.</span>

MORPHOMETRIC CHARACTERISTICS OF A TROPICAL RIVER BASIN, CENTRAL KERALA, INDIA USING GEOSPATIAL TECHNIQUES

The Thutapuzha watershed (TW) is one among the major tributaries of Bharathapuzha – the largest west flowing river in Kerala, India. Morphometric analysis was carried out to determine the spatial variations in the drainage characteristics and to understand the prevailing geologic variation, topographic information and structural setup of TW using Survey of India topographic maps and ASTER-DEM. Geoprocessing techniques has been used for the delineation and calculation of the morphometric parameters of the watershed. The TW sprawled over 1107 Km2 and the study revealed that the watershed includes a sixth order stream and lower order streams mostly dominate the basin with a drianage density of 1.36 m/Km2exhibiting highly resistant subsoil, dense vegetation, and low relief of surface nature. The study indicate that rainfall has a significant role in the drainage development whereas the drainage pattern is controlled by structure and relief. The watershed of TW is moderate to well-drained and exhibited a geomorphic maturity in its physiographic development. The shape parameters revealed the elongated nature of TW having less prone to flood, lower erosion and sediment transport capacities and drainage network development in the watershed. This study strongly brings to light that the drainage morphometric parameters have the enormous potentiality to unveil the hydro-morphologicalcharacteristics of the river basins. Integrating hydro-morphological characteristics with conventional watershed assessment methods would have a beneficial effect on judicious watershed management, which helped to formulate a comprehensive watershed management plan.

A REVIEW ON MARKER-CONTROLLED WATER SHED

The marker-based watershed segmentation algorithm is the main topic of this study. Because the watershed segmentation method over segments the image and produces noise, As a result, numerous academics have put forth various solutions to this challenge, but markercontrolled watersheds are by far the greatest option. Finding the gaps in the body of literature is the main goal of this paper. Reviewing the various segmentation strategies, it was discovered that marker-based segmentation is optimal in most situations since it labels the regions before segmenting them. Yet further research is needed to optimise the marking zones.

Characterisation of the evolution of pore structure with particle breakage during compaction

This work investigated the effect of particle breakage on the pore structure of compacts formed with the discrete element method. The watershed segmentation method was developed to quantify pore properties of the compacts. Results showed the pore size, throat diameter and throat length followed the log-normal distributions for all the compacts except for compacts in which significant particle breakage occurred. With particle breakage, the mean pore size and mean throat diameter decreased exponentially with reducing particle size. The pore structure was also explored through spontaneous interparticle percolation. Results showed that the residence time of the percolating particles followed the log-normal distributions. The coefficient of the radial dispersion increased linearly with depth for all compacts except for a compact with particle breakage. The relations between pore structure and percolation behaviour were established. With decreasing pore size and throat diameter, the mean residence time increased exponentially but the dispersion coefficient had a linear decrease.

Watershed Health Assessment Using GIS and AHP Methods: Application in Stung Sen River Basin, Cambodia

The watershed assessment provides information about the condition of water quality and biological integrity to identify the source of stressors and their impacts. In the present decades, different watershed assessment method has been established to evaluate the cumulative impacts of human activities on watershed health and aquatic systems. This study proposes a new approach for assessing watershed vulnerability to contamination based on spatial analysis using the Geographic Information System (GIS). and Analytic Hierarchy Process (AHP) methods. This new procedure designed to identify vulnerable zones depends on seven basic factors representing watershed characteristics: land use/land cover, sediment load, nitrate load, phosphorus load, soil type, average annual precipitation, and slope. The new watershed vulnerability assessment technique was used to create a map showing the relative vulnerabilities of specific sub-watersheds in the Sen River Basin, the largest sub-basin of the Tonle Sap Lake. The results showed a remarkable difference in watershed susceptibility between the sub-watersheds in their vulnerability to pollution. The approximate area of 10,846 km2 (76%) in the Northwest part and the long distance from the river of the study area were categorized in a range from moderate, low, and very low watershed vulnerability. However, consisting 3,341 km2 (24%) located downstream and near distance from the river were displayed as a very high and high vulnerability to pollution in the watershed. Furthermore, the results of the evaluation of the predictive reliability of the watershed vulnerability assessment method revealed that the proposed approach is suitable as a decision-making tool to predict watershed health. The process of this study indeed provides an application performance for the Sen River Basin and calls for action to sustain the water ecosystem and use.

AFM characterization of physical properties in coal adsorbed with different cations induced by electric pulse fracturing

Pore system in coal is highly heterogeneous, while it is the main occurrence space and transport channel for coalbed methane (CBM). Electric pulse fracturing (EPF) has been considered as an effective approach to improve the coal physical properties for better CBM production. In this work, based on AFM measurement of 40 coal samples collected from Qinshui Basin, we evaluated the physical properties of coal, adsorbed with different cations, after EPF. This was accomplished by first analyzing the breakdown field response process of coal adsorbed with different cations, and then determining the dynamic changes of pore and surface roughness using the watershed method. In the PeakForce QNMTM model, the heterogeneous distribution of the adhesion force with increasing cation valence was quantified. The results show that the ability of medium- and high-rank coals to adsorb cations exhibits in the order of Fe3+ > Ca2+ > K+, attributed to the decrease in the radius of hydrated ions with increasing ionic valence. When the electric field is applied to both ends of coal sample, the plasma channels in coal generate huge energy, resulting in the temperature rise of pore-fracture and throat. Affected by EPF, the porosity of Chengzhuang (CZ) increases from 4.1% to 27.4%, greater than that of Qiyi (QY) from 6.7% to 14.5%. In the surface morphology tests, the surface skewness Rsk of QY adsorbed with different valence cations shifts from positive to negative values, reflecting the change of coal surface height from the right-skewed normal distribution to the left-skewed one. Moreover, the area where the surface height of QY adsorbed cation is below the average value gradually increases with the increase of the chemical valence. For micromechanical properties, the adhesion force of CZ adsorbed with different cations ranges from 0.9 to 20.7 nN, which is less than QY overall. This is mainly due to the weakening of intermolecular forces on the coal surface with the deepening of coalification, resulting in a decrease in adhesion force. Therefore, this study provides new insights into the differences in gas production mechanisms of CBM wells in the same coal seam from different regions.

Cell segmentation from telecentric bright-field transmitted light microscopy images using a Residual Attention U-Net: A case study on HeLa line

Living cell segmentation from bright-field light microscopy images is challenging due to the image complexity and temporal changes in the living cells. Recently developed deep learning (DL)-based methods became popular in medical and microscopy image segmentation tasks due to their success and promising outcomes. The main objective of this paper is to develop a deep learning, U-Net-based method to segment the living cells of the HeLa line in bright-field transmitted light microscopy. To find the most suitable architecture for our datasets, a residual attention U-Net was proposed and compared with an attention and a simple U-Net architecture.The attention mechanism highlights the remarkable features and suppresses activations in the irrelevant image regions. The residual mechanism overcomes with vanishing gradient problem. The Mean-IoU score for our datasets reaches 0.9505, 0.9524, and 0.9530 for the simple, attention, and residual attention U-Net, respectively. The most accurate semantic segmentation results was achieved in the Mean-IoU and Dice metrics by applying the residual and attention mechanisms together. The watershed method applied to this best – Residual Attention – semantic segmentation result gave the segmentation with the specific information for each cell.

Improving visualization of three-dimensional aneurysm features via segmentation with upsampled resolution and gradient enhancement (SURGE)

BackgroundIntracranial aneurysm neck width tends to be overestimated when measured with three-dimensional rotational angiography (3DRA) compared with two-dimensional digital subtraction angiography (2D-DSA), owing to high curvature at the neck. This...

Quantifying Raveling Using 3D Technology with Loss of Aggregates as a New Performance Indicator

Pavement raveling is one of the predominant distresses in the United States that impacts roadway safety and driver comfort on open-graded friction course (OGFC) pavements. Raveling specific treatments, such as fog seal and micro-milling the OGFC layer, can prolong pavement life and reduce resurfacing costs and environmental impact. However, with the current qualitative condition assessment methods (which rate pavements at Severity Levels 1–3 or as light, moderate, or severe), it is difficult to determine the optimal timing for these raveling treatments to be most effective. Therefore, there is an urgent need to develop a method to quantitatively evaluate the raveling condition. While 3D pavement technology provides opportunities for quantifying pavement raveling conditions using 3D pavement surface data, there are two main challenges for quantifying pavement raveling: (1) estimating a reference surface that represents the pavement without any raveling so that the actual pavement can be compared to the reference surface to quantify the raveling, and (2) obtaining pavement images with quantified raveling conditions (aggregate loss volume) for validation. This paper proposes a method with the loss of aggregate as a new performance indicator to automatically quantify raveling using 3D pavement surface data already collected by transportation agencies for pavement evaluation. The proposed method is validated using pavement images (with known aggregate loss) from simulated pavement mats fabricated in the lab and synthetic pavement images obtained by procedural generation. The proposed method consists of (1) 3D data acquisition; (2) pre-processing with (a) outlier removal and image smoothing, (b) two-sensor image stitching, and (c) range image rectification; (3) raveling detection using (a) region of interest selection, (b) reference surface estimation, (c) potential aggregate loss identification, and (d) noise removal; and (4) aggregate loss quantification. The validation results show a strong correlation (R = 0.99) between the computed aggregate loss and the expected aggregate loss. Better performance was observed with the proposed method than with other methods (such as the watershed method and the model fitting method). The proposed method provides a cost-effective means to quantify the loss of aggregates in support of quantitative raveling condition forecasting by leveraging 3D pavement data already collected by transportation agencies.

Individual Tree Crown Segmentation and Crown Width Extraction From a Heightmap Derived From Aerial Laser Scanning Data Using a Deep Learning Framework.

Deriving individual tree crown (ITC) information from light detection and ranging (LiDAR) data is of great significance to forest resource assessment and smart management. After proof-of-concept studies, advanced deep learning methods have been shown to have high efficiency and accuracy in remote sensing data analysis and geoscience problem solving. This study proposes a novel concept for synergetic use of the YOLO-v4 deep learning network based on heightmaps directly generated from airborne LiDAR data for ITC segmentation and a computer graphics algorithm for refinement of the segmentation results involving overlapping tree crowns. This concept overcomes the limitations experienced by existing ITC segmentation methods that use aerial photographs to obtain texture and crown appearance information and commonly encounter interference due to heterogeneous solar illumination intensities or interlacing branches and leaves. Three generative adversarial networks (WGAN, CycleGAN, and SinGAN) were employed to generate synthetic images. These images were coupled with manually labeled training samples to train the network. Three forest plots, namely, a tree nursery, forest landscape and mixed tree plantation, were used to verify the effectiveness of our approach. The results showed that the overall recall of our method for detecting ITCs in the three forest plot types reached 83.6%, with an overall precision of 81.4%. Compared with reference field measurement data, the coefficient of determination R2 was ≥ 79.93% for tree crown width estimation, and the accuracy of our deep learning method was not influenced by the values of key parameters, yielding 3.9% greater accuracy than the traditional watershed method. The results demonstrate an enhancement of tree crown segmentation in the form of a heightmap for different forest plot types using the concept of deep learning, and our method bypasses the visual complications arising from aerial images featuring diverse textures and unordered scanned points with irregular geometrical properties.

Image Segmentation via Multiscale Perceptual Grouping

The human eyes observe an image through perceptual units surrounded by symmetrical or asymmetrical object contours at a proper scale, which enables them to quickly extract the foreground of the image. Inspired by this characteristic, a model combined with multiscale perceptual grouping and unit-based segmentation is proposed in this paper. In the multiscale perceptual grouping part, a novel total variation regularization is proposed to smooth the image into different scales, which removes the inhomogeneity and preserves the edges. To simulate perceptual units surrounded by contours, the watershed method is utilized to cluster pixels into groups. The scale of smoothness is determined by the number of perceptual units. In the segmentation part, perceptual units are regarded as the basic element instead of discrete pixels in the graph cut. The appearance models of the foreground and background are constructed by combining the perceptual units. According to the relationship between perceptual units and the appearance model, the foreground can be segmented through a minimum-cut/maximum-flow algorithm. The experiment conducted on the CMU-Cornell iCoseg database shows that the proposed model has a promising performance.

Two-wavelength image detection of early decayed oranges by coupling spectral classification with image processing

Citrus decay is one of the most serious postharvest diseases, which can cause great economic losses and food safety problem. Early decay has no obvious features, which makes rapid detection and grading of decayed citrus fruit a major challenge for citrus industry. This study utilized the unique image-spectrum fusion property of hyperspectral imaging, and proposed a strategy to quickly identify the citrus with early decay using only two wavelength images. The typical average spectra of sound and decayed tissues were extracted. The linear partial least squares-discriminant analysis (PLS-DA) and nonlinear back-propagation artificial neural network (BP-ANN) models were constructed for classifying two types of tissues. MC-UVE-SPA algorithm by combining Monte Carlo cross-validation (MC-UVE) with successive projections algorithm (SPA) was used to extract 16 variables characterizing two types of tissues. The wavelength images corresponding to the extracted variables were performed principal component analysis (PCA) to find the optimal PC image. Only two wavelength images at 568.8 nm and 771.2 nm were selected by analyzing the weighting coefficients of the third principal component (PC3) image. An improved watershed segmentation method was proposed to segment decay region in oranges based on PC2 image of the selected two wavelength images. Classification performance of the proposed algorithm was evaluated by all samples. The results showed that the overall classification accuracy of 96.6% was achieved, with 100% and 91.3% for the decayed and sound oranges respectively. The proposed detection strategy only involves two wavelength images, which will contribute to the establishment of a fast and low-cost multispectral imaging system for detection of citrus with early decay.

Breast cancer histopathological images recognition based on two-stage nuclei segmentation strategy.

Pathological examination is the gold standard for breast cancer diagnosis. The recognition of histopathological images of breast cancer has attracted a lot of attention in the field of medical image processing. In this paper, on the base of the Bioimaging 2015 dataset, a two-stage nuclei segmentation strategy, that is, a method of watershed segmentation based on histopathological images after stain separation, is proposed to make the dataset recognized to be the carcinoma and non-carcinoma recognition. Firstly, stain separation is performed on breast cancer histopathological images. Then the marker-based watershed segmentation method is used for images obtained from stain separation to achieve the nuclei segmentation target. Next, the completed local binary pattern is used to extract texture features from the nuclei regions (images after nuclei segmentation), and color features were extracted by using the color auto-correlation method on the stain-separated images. Finally, the two kinds of features were fused and the support vector machine was used for carcinoma and non-carcinoma recognition. The experimental results show that the two-stage nuclei segmentation strategy proposed in this paper has significant advantages in the recognition of carcinoma and non-carcinoma on breast cancer histopathological images, and the recognition accuracy arrives at 91.67%. The proposed method is also applied to the ICIAR 2018 dataset to realize the automatic recognition of carcinoma and non-carcinoma, and the recognition accuracy arrives at 92.50%.

Landslide Susceptibility Mapping along a Rapidly Uplifting River Valley of the Upper Jinsha River, Southeastern Tibetan Plateau, China

As a result of the influence of plate movement, the upper reaches of Jinsha River have strong geological tectonic activities, large topographic fluctuations, and complex climate characteristics, which result in the frequent occurrence of landslide disasters. Hence, there is the need to carry out landslide susceptibility mapping in the upper reaches of Jinsha River to ensure the safety of local people’s property and the safe exploitation of hydraulic resources. In this study, InSAR technology and a field geological survey were used to map the landslides. Then, the curvature watershed method was used to divide the slope units. A conditioning factor system was established, which can reflect the characteristics of the rapid uplift and vertical distribution of rainfall in the special geological environment of the study area. Finally, logistic regression, random forest, and artificial neural network models were used to establish the landslide susceptibility model. The results show that the random forest model is optimal for the landslide susceptibility mapping in this area. Additionally, the area percentages of the very low, low, moderate, high, and very high susceptibility classes were 40.13%, 20.06%, 13.39%, 12.55%, and 13.87%, respectively. Based on the analysis of the landslide susceptibility map, we suggest that the landslide geological hazards resulting from the rapid uplift of the Tibetan Plateau and the significant decrease in sea level during a glacial period in the upper reaches of Jinsha River are controlled by the double disaster effect of the geodynamic system. Consequently, this study can guide local prevention and mitigation.

Role of Segmentation in Lung Nodules CT Scan Images for High Performance: Case of Recent Findings

Early stage of lung cancer detection is being very vital in today’s scenario. Several promising technological diagnostic automated tools developed and are used to predict the patient's survival using intelligent lung nodules analysis in Computed Tomography (CT) images. Recently, the advancement of human health diagnostic technologies, Computer-Aided Detection Systems (CADe) are developed diurnal rhythms to provide higher accuracy and better performance rate. In this research work, we focuses on a documenting the automatic segmentation of lung cancer nodules by various methods developed in recent years of the timeline with respect to the realization. The proposed work is criterion based objectives to establish the performance measures and to propose new solution for Lung nodules segmentation and detection on CT Scanned digital images, which is the fundamental and epochal measure to attain the high level of performance. In analyzing each of these technical execution on detection on lung nodule, and experimented to 4 different kinds of lung nodules CT scanned digital images. The phenomenon show segmentation methods involving convolution neural networks were better when compared to other existing methods. U-Net method gave improved accuracy. Dice score 0.981±0.009 and sensitivity 0.994±0.002 are maximum by using the method Dense R2UNet. F1-score 97% is best achieved by Cascade Convolutional Network. Recall rate is achieved to the maximum 99.1% through Improved 3D-UNet Neural Network. Precision 0.982±0.009 and accuracy 0.988±0.018 are best attained by Dense R2UNet. In the case, semi-transparent nodules segmentation, Watershed method found to be highly appropriate select. Watershed arse nodules segment correspond well to vessels and semi-transparent nodules, and exhibit low sensitivity in solitary or lone nodules. The higher rate of efficient segmentation is directly proportional to higher rate of performance. Keywords: CNN, Lung Cancer, CT Image Segmentation, Watershed, R2Unet.

A Study on Region-Based Image Segmentation Techniques

This paper addresses the comparison of two popular region-based image segmentation techniques namely the Watershed method and the Mean-shift method. The watershed method is a mathematical morphology based image segmentation approach while Mean-shift method is a data-clustering method that searches for the local maximal density points and then groups all the data to the clusters defined by these maximal density points. Here the efficiency of the both segmentation techniques is presented

Application Of Dynamic Segmentation In Stroke Detection Software With ANN

One way to find out whether there is a stroke is to do a CT scan . But the results of the examination with a new CT scan can be obtained in quite a long time. In addition, sometimes there are differences of opinion between doctors and radiologists regarding what is seen from the results of the examination. This research was conducted to produce a software that can later be integrated with the existing system on the CT Scan tool so that it can immediately be known whether or not stroke is present from the CT Scan results. In this study, a dynamic image segmentation method is implemented, namely the watershed transformation method which will later produce regions as a feature for the stroke detection process carried out with the backpropagation algorithm. From experiments conducted on CT scan images of the brain, this method can detect stroke well. The results obtained are 100% for training data and 90% for test data.