Accurate Segmentation Algorithm Research Articles

A Precise Segmentation Algorithm of Pumpkin Seedling Point Cloud Stem Based on CPHNet.

Accurate segmentation of the stem of pumpkin seedlings has a great influence on the modernization of pumpkin cultivation, and can provide detailed data support for the growth of pumpkin plants. We collected and constructed a pumpkin seedling point cloud dataset for the first time. Potting soil and wall background in point cloud data often interfere with the accuracy of partial cutting of pumpkin seedling stems. The stem shape of pumpkin seedlings varies due to other environmental factors during the growing stage. The stem of the pumpkin seedling is closely connected with the potting soil and leaves, and the boundary of the stem is easily blurred. These problems bring challenges to the accurate segmentation of pumpkin seedling point cloud stems. In this paper, an accurate segmentation algorithm for pumpkin seedling point cloud stems based on CPHNet is proposed. First, a channel residual attention multilayer perceptron (CRA-MLP) module is proposed, which suppresses background interference such as soil. Second, a position-enhanced self-attention (PESA) mechanism is proposed, enabling the model to adapt to diverse morphologies of pumpkin seedling point cloud data stems. Finally, a hybrid loss function of cross entropy loss and dice loss (HCE-Dice Loss) is proposed to address the issue of fuzzy stem boundaries. The experimental results show that CPHNet achieves a 90.4% average cross-to-merge ratio (mIoU), 93.1% average accuracy (mP), 95.6% average recall rate (mR), 94.4% F1 score (mF1) and 0.03 plants/second (speed) on the self-built dataset. Compared with other popular segmentation models, this model is more accurate and stable for cutting the stem part of the pumpkin seedling point cloud.

Synergizing a Deep Learning and Enhanced Graph-Partitioning Algorithm for Accurate Individual Rubber Tree-Crown Segmentation from Unmanned Aerial Vehicle Light-Detection and Ranging Data

The precise acquisition of phenotypic parameters for individual trees in plantation forests is important for forest management and resource exploration. The use of Light-Detection and Ranging (LiDAR) technology mounted on Unmanned Aerial Vehicles (UAVs) has become a critical method for forest resource monitoring. Achieving the accurate segmentation of individual tree crowns (ITCs) from UAV LiDAR data remains a significant technical challenge, especially in broad-leaved plantations such as rubber plantations. In this study, we designed an individual tree segmentation framework applicable to dense rubber plantations with complex canopy structures. First, the feature extraction module of PointNet++ was enhanced to precisely extract understory branches. Then, a graph-based segmentation algorithm focusing on the extracted branch and trunk points was designed to segment the point cloud of the rubber plantation. During the segmentation process, a directed acyclic graph is constructed using components generated through grey image clustering in the forest. The edge weights in this graph are determined according to scores calculated using the topologies and heights of the components. Subsequently, ITC segmentation is performed by trimming the edges of the graph to obtain multiple subgraphs representing individual trees. Four different plots were selected to validate the effectiveness of our method, and the widths obtained from our segmented ITCs were compared with the field measurement. As results, the improved PointNet++ achieved an average recall of 94.6% for tree trunk detection, along with an average precision of 96.2%. The accuracy of tree-crown segmentation in the four plots achieved maximal and minimal R2 values of 98.2% and 92.5%, respectively. Further comparative analysis revealed that our method outperforms traditional methods in terms of segmentation accuracy, even in rubber plantations characterized by dense canopies with indistinct boundaries. Thus, our algorithm exhibits great potential for the accurate segmentation of rubber trees, facilitating the acquisition of structural information critical to rubber plantation management.

A Grape Dataset for Instance Segmentation and Maturity Estimation

Grape maturity estimation is vital in precise agriculture as it enables informed decision making for disease control, harvest timing, grape quality, and quantity assurance. Despite its importance, there are few large publicly available datasets that can be used to train accurate and robust grape segmentation and maturity estimation algorithms. To this end, this work proposes the CERTH grape dataset, a new sizeable dataset that is designed explicitly for evaluating deep learning algorithms in grape segmentation and maturity estimation. The proposed dataset is one of the largest currently available grape datasets in the literature, consisting of around 2500 images and almost 10 k grape bunches, annotated with masks and maturity levels. The images in the dataset were captured under various illumination conditions and viewing angles and with significant occlusions between grape bunches and leaves, making it a valuable resource for the research community. Thorough experiments were conducted using a plethora of general object detection methods to provide a baseline for the future development of accurate and robust grape segmentation and maturity estimation algorithms that can significantly advance research in the field of viticulture.

Soybean-MVS: Annotated Three-Dimensional Model Dataset of Whole Growth Period Soybeans for 3D Plant Organ Segmentation

The study of plant phenotypes based on 3D models has become an important research direction for automatic plant phenotype acquisition. Building a labeled three-dimensional dataset of the whole growth period can help the development of 3D crop plant models in point cloud segmentation. Therefore, the demand for 3D whole plant growth period model datasets with organ-level markers is growing rapidly. In this study, five different soybean varieties were selected, and three-dimensional reconstruction was carried out for the whole growth period (13 stages) of soybean using multiple-view stereo technology (MVS). Leaves, main stems, and stems of the obtained three-dimensional model were manually labeled. Finally, two-point cloud semantic segmentation models, RandLA-Net and BAAF-Net, were used for training. In this paper, 102 soybean stereoscopic plant models were obtained. A dataset with original point clouds was constructed and the subsequent analysis confirmed that the number of plant point clouds was consistent with corresponding real plant development. At the same time, a 3D dataset named Soybean-MVS with labels for the whole soybean growth period was constructed. The test result of mAccs at 88.52% and 87.45% verified the availability of this dataset. In order to further promote the study of point cloud segmentation and phenotype acquisition of soybean plants, this paper proposed an annotated three-dimensional model dataset for the whole growth period of soybean for 3D plant organ segmentation. The release of the dataset can provide an important basis for proposing an updated, highly accurate, and efficient 3D crop model segmentation algorithm. In the future, this dataset will provide important and usable basic data support for the development of three-dimensional point cloud segmentation and phenotype automatic acquisition technology of soybeans.

Accurate segmentation algorithm of acoustic neuroma in the cerebellopontine angle based on ACP-TransUNet.

Acoustic neuroma is one of the most common tumors in the cerebellopontine angle area. Patients with acoustic neuroma have clinical manifestations of the cerebellopontine angle occupying syndrome, such as tinnitus, hearing impairment and even hearing loss. Acoustic neuromas often grow in the internal auditory canal. Neurosurgeons need to observe the lesion contour with the help of MRI images, which not only takes a lot of time, but also is easily affected by subjective factors. Therefore, the automatic and accurate segmentation of acoustic neuroma in cerebellopontine angle on MRI is of great significance for surgical treatment and expected rehabilitation. In this paper, an automatic segmentation method based on Transformer is proposed, using TransUNet as the core model. As some acoustic neuromas are irregular in shape and grow into the internal auditory canal, larger receptive fields are thus needed to synthesize the features. Therefore, we added Atrous Spatial Pyramid Pooling to CNN, which can obtain a larger receptive field without losing too much resolution. Since acoustic neuromas often occur in the cerebellopontine angle area with relatively fixed position, we combined channel attention with pixel attention in the up-sampling stage so as to make our model automatically learn different weights by adding the attention mechanism. In addition, we collected 300 MRI sequence nuclear resonance images of patients with acoustic neuromas in Tianjin Huanhu hospital for training and verification. The ablation experimental results show that the proposed method is reasonable and effective. The comparative experimental results show that the Dice and Hausdorff 95 metrics of the proposed method reach 95.74% and 1.9476 mm respectively, indicating that it is not only superior to the classical models such as UNet, PANet, PSPNet, UNet++, and DeepLabv3, but also show better performance than the newly-proposed SOTA (state-of-the-art) models such as CCNet, MANet, BiseNetv2, Swin-Unet, MedT, TransUNet, and UCTransNet.

Pulmonary Nodule Segmentation Framework Based on Fine-Tuned and Pretrained Deep Neural Network Using CT Images

Deep learning is one of the most rapidly growing and emerging technologies in pulmonary nodule segmentation. However, the different shape, size, and location of the nodules make it extremely difficult to segment correctly. The purpose of this study is to obtain a fast and accurate segmentation algorithm with less number of stages. A fine-tuned dual skip connection-based segmentation framework is proposed that integrates pretrained residual neural network (ResNet) 152 with the U-Net architecture, namely, ResiU-Net. Nine different pretrained and fine-tuned encoder backbones, such as ResNet18, ResNet 34, ResNet 50, ResNet 101, ResNet 152, SEResNet18, SE-ResNet34, ResNext 101, and ResNext 50 are compared and the proposed ResiU-Net approach gives the best results. Also, the fine-tuned ResiU-Net performs better than nontuned ResiU-Net. 1224 computed tomography patient images with different nodule shapes and sizes are selected. The proposed method achieves 97.44% F score, 95.02% intersection over union score, 94.87% dice score, 0.34% binary cross-entropy loss, and 0.7585 combined dice coefficient and binary focal loss. The proposed ResiU-Net outperforms the state-of-the-art methods and reports the best evaluation metrics. The time taken by the model to train is 43 min. Hence, the proposed model is a fast and accurate segmentation approach.

SEMANTIC SEGMENTATION OF UAV IMAGES BASED ON U-NET IN URBAN AREA

Abstract. Semantic segmentation of aerial data has been one of the leading researches in the field of photogrammetry, remote sensing, and computer vision in recent years. Many applications, including airborne mapping of urban scenes, object positioning in aerial images, automatic extraction of buildings from remote sensing or high-resolution aerial images, etc., require accurate and efficient segmentation algorithms. According to the high potential of deep learning algorithms in the classification of complex scenes, this paper aims to train a deep learning model to evaluate the semantic segmentation accuracy of UAV-based images in urban areas. The proposed method implements a deep learning framework based on the U-Net encoder-decoder architecture, which extracts and classifies features through layers of convolution, max pooling, activation, and concatenation in an end-to-end process. The obtained results compare with two traditional machine learning models, Random Forest (RF) and Multi-Layer Perceptron (MLP). They rely on two steps that involve extracting features and classifying images. In this study, the experiments are performed on the UAVid2020 semantic segmentation dataset from the ISPRS database. Results show the effectiveness of the proposed deep learning framework, so that the U-Net architecture achieved the best results with 75.15% overall accuracy, compared to RF and MLP algorithms with 52.51% and 54.65% overall accuracy, respectively.

HMNet: Hierarchical Multi-Scale Brain Tumor Segmentation Network.

An accurate and efficient automatic brain tumor segmentation algorithm is important for clinical practice. In recent years, there has been much interest in automatic segmentation algorithms that use convolutional neural networks. In this paper, we propose a novel hierarchical multi-scale segmentation network (HMNet), which contains a high-resolution branch and parallel multi-resolution branches. The high-resolution branch can keep track of the brain tumor's spatial details, and the multi-resolution feature exchange and fusion allow the network's receptive fields to adapt to brain tumors of different shapes and sizes. In particular, to overcome the large computational overhead caused by expensive 3D convolution, we propose a lightweight conditional channel weighting block to reduce GPU memory and improve the efficiency of HMNet. We also propose a lightweight multi-resolution feature fusion (LMRF) module to further reduce model complexity and reduce the redundancy of the feature maps. We run tests on the BraTS 2020 dataset to determine how well the proposed network would work. The dice similarity coefficients of HMNet for ET, WT, and TC are 0.781, 0.901, and 0.823, respectively. Many comparative experiments on the BraTS 2020 dataset and other two datasets show that our proposed HMNet has achieved satisfactory performance compared with the SOTA approaches.

Automatic retrieval of volcanic SO2 emission source from TROPOMI products

Volcanic sulfur dioxide (SO2) satellite observations are key for monitoring volcanic activity, and for mitigation of the associated risks on both human health and aviation safety. Automatic analysis of this data source, including robust source emission retrieval, is in turn essential for near real-time monitoring applications. We have developed fast and accurate SO2 plume classifier and segmentation algorithms using classic clustering, segmentation and image processing techniques. These algorithms, applied to measurements from the TROPOMI instrument onboard the Sentinel-5 Precursor platform, can help in the accurate source estimation of volcanic SO2 plumes originating from various volcanoes. In this paper, we demonstrate the ability of different pixel classification methodologies to retrieve SO2 source emission with a good accuracy. We compare the algorithms, their strengths and shortcomings, and present plume classification results for various active volcanoes throughout the year 2021, including examples from Etna (Italy), Sangay and Reventador (Ecuador), Sabancaya and Ubinas (Peru), Scheveluch and Klyuchevskoy (Russia), as well as Ibu and Dukono (Indonesia). The developed algorithms, shared as open-source code, contribute to improving analysis and monitoring of volcanic emissions from space.

RU-Net2+: A Deep Learning Algorithm for Accurate Brain Tumor Segmentation and Survival Rate Prediction

RU-Net2+: A Deep Learning Algorithm for Accurate Brain Tumor Segmentation and Survival Rate Prediction

Fast and accurate superpixel segmentation algorithm with a guidance image

Superpixels are generated by automatically grouping image pixels into lots and lots of compact segments. In computer vision, it is widely used as an effective way to reduce the number of image primitives for subsequent tasks and for recognizing objects’ contours due to its excellent boundary adhesion. The primary concerns of a superpixel generation algorithm are its efficiency and accuracy. In this document, we aim to propose a fast and precise superpixel segmentation algorithm with a guidance image. Specifically, we introduce a rolling filter that can remove details while well preserving the boundaries to obtain the guidance image. In the meantime, we introduce a robust edge confidence operator to accurately detect image boundaries. From this, we define a distance measurement with adaptive parameters for each image. In this way, we can adapt and accurately group pixels into regions according to the new distance measurement. Furthermore, we adopt a noniterative framework to generate superpixels in real time by processing all pixels once. The experimental results show that the proposed methodology achieves state-of-the-art performance on two sets of reference data while operating in real time.

Clinical tooth segmentation based on local enhancement.

The tooth arrangements of human beings are challenging to accurately observe when relying on dentists’ naked eyes, especially for dental caries in children, which is difficult to detect. Cone-beam computer tomography (CBCT) is used as an auxiliary method to measure patients’ teeth, including children. However, subjective and irreproducible manual measurements are required during this process, which wastes much time and energy for the dentists. Therefore, a fast and accurate tooth segmentation algorithm that can replace repeated calculations and annotations in manual segmentation has tremendous clinical significance. This study proposes a local contextual enhancement model for clinical dental CBCT images. The local enhancement model, which is more suitable for dental CBCT images, is proposed based on the analysis of the existing contextual models. Then, the local enhancement model is fused into an encoder–decoder framework for dental CBCT images. At last, extensive experiments are conducted to validate our method.

Two fully automated data-driven 3D whole-breast segmentation strategies in MRI for MR-based breast density using image registration and U-Net with a focus on reproducibility

Presence of higher breast density (BD) and persistence over time are risk factors for breast cancer. A quantitatively accurate and highly reproducible BD measure that relies on precise and reproducible whole-breast segmentation is desirable. In this study, we aimed to develop a highly reproducible and accurate whole-breast segmentation algorithm for the generation of reproducible BD measures. Three datasets of volunteers from two clinical trials were included. Breast MR images were acquired on 3 T Siemens Biograph mMR, Prisma, and Skyra using 3D Cartesian six-echo GRE sequences with a fat-water separation technique. Two whole-breast segmentation strategies, utilizing image registration and 3D U-Net, were developed. Manual segmentation was performed. A task-based analysis was performed: a previously developed MR-based BD measure, MagDensity, was calculated and assessed using automated and manual segmentation. The mean squared error (MSE) and intraclass correlation coefficient (ICC) between MagDensity were evaluated using the manual segmentation as a reference. The test-retest reproducibility of MagDensity derived from different breast segmentation methods was assessed using the difference between the test and retest measures (Δ2-1), MSE, and ICC. The results showed that MagDensity derived by the registration and deep learning segmentation methods exhibited high concordance with manual segmentation, with ICCs of 0.986 (95%CI: 0.974-0.993) and 0.983 (95%CI: 0.961-0.992), respectively. For test-retest analysis, MagDensity derived using the registration algorithm achieved the smallest MSE of 0.370 and highest ICC of 0.993 (95%CI: 0.982-0.997) when compared to other segmentation methods. In conclusion, the proposed registration and deep learning whole-breast segmentation methods are accurate and reliable for estimating BD. Both methods outperformed a previously developed algorithm and manual segmentation in the test-retest assessment, with the registration exhibiting superior performance for highly reproducible BD measurements.

Impact crack quantification analyses in 3-D angle-interlock woven composite using image segmentation method

Crack distribution quantifications are key to predict residual strength of composite materials after impact. Here we report impact crack distribution quantifications of 3-D angle-interlock woven composites (3-DAWCs) which were observed with micro-computed tomography (micro-CT), and also characterized with finite element analyses (FEA). A fast and accurate image segmentation algorithm has been employed to identify the inside crack distributions. A damage variable has been adopted to quantify the crack propagation. We found that the weft yarn with straight arrangement is conducive to stress propagation, and is the main load carrier. The large deflection and buckling structure of warp yarns inhibit the stress propagation on the back face. It is shown that the cohesive damages locate at the interlacing points and the interface between resin and surface warp yarns after the first three successive impacts. The FEA model has been verified with the test data for predicting the damage initiation, evolution and failure mode of materials under impact, which is of great significance for the 3-D woven composite material design.

An accurate detection and segmentation model of obscured green fruits

Accurate detection and segmentation of obscured green fruits is a key factor in the efficiency of the harvesting robot. In the real and complex orchard environment, there are many influencing factors such as illumination changes, occlusion of branches and leaves, weather, etc., which bring difficulties to the work of picking robots. Therefore, higher requirements are put forward for accurate detection and segmentation algorithms of machine vision. An accurate detection and segmentation method for obscured green fruit based on a fully convolutional one-stage (FCOS) object detection model is proposed. First, the feature pyramid network (FPN) in FCOS is replaced by residual feature pyramid network (RFPN), which improves the detection accuracy of green fruits with different sizes. Next, as the boundaries of fuzzy fruits are often not accurately identified, a two-layer convolutional block attention network is connected to a modified FCOS to recover the edges of incomplete green fruits. The experimental results show that the detection and segmentation accuracy of green fruit reaches 81.2% and 85.3% on Apple dataset. On the Apple-ape dataset, the detection accuracy and segmentation accuracy reached 77.2% and 79.7%, respectively. The Params of the model is 39.7 M, and the floating point operations (FLOPs) are 169.8G, which has great advantages over other ten advanced models in terms of accuracy and complexity. The model also improves the working efficiency of orchard harvesting robots and lays the foundation for the further development of intelligent agriculture.

Tubular shape aware data generation for segmentation in medical imaging.

Chest X-ray is one of the most widespread examinations of the human body. In interventional radiology, its use is frequently associated with the need to visualize various tube-like objects, such as puncture needles, guiding sheaths, wires, and catheters. Detection and precise localization of these tube-like objects in the X-ray images are, therefore, of utmost value, catalyzing the development of accurate target-specific segmentation algorithms. Similar to the other medical imaging tasks, the manual pixel-wise annotation of the tubes is a resource-consuming process. In this work, we aim to alleviate the lack of annotated images by using artificial data. Specifically, we present an approach for synthetic generation of the tube-shaped objects, with a generative adversarial network being regularized with a prior-shape constraint. Namely, our model uses Frangi-based regularization to draw synthetic tubes in the predefined fake mask regions and, then, uses the adversarial component to preserve the global realistic appearance of the synthesized image. Our method eliminates the need for the paired image-mask data and requires only a weakly labeled dataset, with fine-tuning on a small paired sample (10-20 images) proving sufficient to reach the accuracy of the fully supervised models. We report the applicability of the approach for the task of segmenting tubes and catheters in the X-ray images, whereas the results should also hold for the other acquisition modalities and image computing applications that contain tubular objects.

3-D Prostate MR and TRUS Images Detection and Segmentation for Puncture Biopsy

Prostate cancer is one of the male cancers with high mortality rate in the world. The mainstream diagnostic method is puncture biopsy. Accurate detection and segmentation of the prostate in MR (Magnetic Resonance) and TRUS (Transrectal Ultrasonography) images is an important prerequisite for image-guided puncture biopsy. Manual or semi-automatic detection and segmentation methods and traditional automatic detection and segmentation methods are difficult to achieve real-time and accurate detection and segmentation under the conditions of poor imaging effect or less labeled data. In this paper, a real-time and accurate automatic detection and segmentation algorithm DRC U-Net for 3D MR and TRUS images of prostate is proposed for prostate puncture biopsy. A novel convolution module that integrates residual connections, dense connections, and deep separable convolutions is proposed for multi-scale feature extraction and fusion. At the same time, deep supervision mechanism and a variety of attention mechanisms are integrated in the network to improve training efficiency and ensure segmentation effect. The experimental results on the public MR dataset Promise12 and the private TRUS dataset show that the real-time performance and accuracy of this method are better than the existing 3D medical image detection and segmentation methods, such as 3D U-Net and V-Net, and also have some advantages compared with the most advanced 3D medical image detection and segmentation methods, which proves that the proposed network structure has important clinical significance and practical value for image-guided puncture biopsy.

Two-stage multitask U-Net construction for pulmonary nodule segmentation and malignancy risk prediction.

Accurate segmentation of pulmonary nodules is important for image-driven nodule analysis and nodule malignancy risk prediction. However, due to interobserver variability caused by manual segmentation, an accurate and robust automatic segmentation method has become an essential task. Therefore, the aim of the present study was to construct an accurate segmentation and malignant risk prediction algorithm for pulmonary nodules. In the present study, we proposed a coarse-to-fine 2-stage framework consisting of the following 2 convolutional neural networks: a 3D multiscale U-Net used for localization and a 2.5D multiscale separable U-Net (MSU-Net) used for segmentation refinement. A multitask framework was proposed for nodules' malignancy risk prediction. Features from encoding and decoding paths of MSU-Net were integrated for pathology or morphology characteristic classification. Experimental results showed that our method achieved state-of-art results on the Lung Image Database Consortium and Image Database Resource Initiative dataset. The proposed method achieved a Dice similarity coefficient (DSC) of 83.04% and an overlapping error of 27.47% on the dataset. Our method achieved accuracy of 77.8% and area under the receiver-operating characteristic curve of 84.3% for malignancy risk prediction. Moreover, we compared our method with the inter-radiologist agreement, and the average DSC difference was only 0.39%. The results showed the effectiveness of the multitask end-to-end framework. The coarse-to-fine 2.5D strategy increased the accuracy and efficiency of pulmonary nodule segmentation and malignancy risk prediction of the computer-aided diagnosis system. In clinical practice, doctors can obtain accurate morphological characteristics and quantitative information of nodules by using the proposed method, so as to make future treatment plan.

Accurate Stereo-Vision-Based Flying Droplet Volume Measurement Method

Accurate volume measurement of flying droplets in the Mura-free manufacturing of inkjet printed organic light-emitting diodes (IJP OLEDs) remains a challenge due to the micrometer scale and high speeds of the droplets. Most existing vision-based methods pay little attention to image segmentation and estimate the droplet volume based on a symmetry assumption, which yields poor accuracy, especially for droplets with an unknown shape. In this article, an accurate measurement method for flying droplet volume based on stereo vision is proposed. First, an image acquisition technique based on stereo vision and synchronous triggering is used to obtain multiview high-resolution droplet images. Then, an accurate binocular droplet image segmentation algorithm is proposed to segment the images with different degrees of blur to achieve similar accuracy. Finally, a polar-coordinate-Hermite-interpolation-based droplet reconstruction (PHDR) algorithm is proposed to reconstruct droplets and to calculate volume on the basis of the contours obtained by the binocular droplet image segmentation algorithm without any symmetry assumptions. The droplet measurement system (DMS) is set up, and a series of experiments are conducted to verify the performance of the proposed method. The results show the advantages of our algorithms, that the volume measurement accuracy is ±3%, and that the relative standard uncertainty is better than 2.1%.

A Classification-Guided Segmentation Algorithm Based on Deep Learning for Epithelium Segmentation in Histopathological Images of Radicular Cysts.

In histopathological analysis of radicular cysts (RCs), lesions in epithelium can provide pathologists with rich information on pathologic degree, which is helpful to determine the type of periapical lesions and make precise treatment planning. Automatic segmentation and localization of epithelium from whole slide images (WSIs) can assist pathologists to complete pathological diagnosis more quickly. However, the class imbalance problem caused by the small proportion of fragmented epithelium in RCs imposes challenge on the typical automatic one-stage segmentation method. In this paper, we proposed a classification-guided segmentation algorithm (CGSA) for accurate segmentation. Our method was a two-stage model, including a classification network for region of interest (ROI) location and a segmentation network guided by classification. The classification stage eliminated most irrelevant areas and alleviated the class imbalance problem faced by the segmentation model. The results of 5-fold cross validation demonstrated that CGSA outperformed the one-stage segmentation method which was lacking in prior epithelium localization information. The epithelium segmentation achieved an overall Dice's coefficient of 0.722, and intersection over union (IoU) of 0.593, which improved by 5.5% and 5.9% respectively compared with the one-stage segmentation method using UNet.Clinical Relevance- This work presents a framework for automatic epithelium segmentation in histopathological images of RCs. It can be applied to make up for the shortcomings of manual annotation which is labor-intensive, time-consuming and objective.