Preliminary Segmentation Results Research Articles

WITHDRAWN: Coronary artery segmentation based on ACMA-Net and unscented Kalman filter algorithm

Accurate segmentation of coronary arteries is a prerequisite for high-precision reconstruction of three-dimensional (3D) models of coronary arteries. With personalized 3D coronary anatomy, clinicians can more readily visually determine the location, type, and severity of a stenotic lesion. However, due to the complex structure and small size of coronary arteries, and the interference of image acquisition conditions, 3D models of segmented and reconstructed coronary arteries are noisy and prone to disconnection; accurate segmentation of coronary arteries is still a challenge.To overcome the above challenges, this study proposes a two-stage coronary segmentation algorithm. In the first stage, we added an atrous convolution feature fusion module (ACFFM) and a multi-axis attention module (MAM) to 3D U-Net, referred to as ACMA-Net, to enhance the feature expression ability of the network and effect the preliminary segmentation of coronary arteries. In the second stage, the preliminary segmentation results were skeletonized and endpoint detection was performed. The regions of coronary artery disconnection were determined by finding the minimum distance between the main trunk branch and the endpoint of each disconnected branch. The disconnected skeleton was repaired by reconnection using an unscented Kalman filter (UKF) algorithm.On the private CORO dataset and the public ASOCA dataset, our method obtains an average Dice coefficient of 0.818 and 0.731, respectively, so it outperforms most of the state-of-the-art algorithms compared, proving its excellent effectiveness and robustness.

Advanced integrated segmentation approach for semi-supervised infrared ship target identification

Infrared ship target segmentation is the important basis of infrared guided weapon in the sea-air context. Typically, accurate infrared ship target segmentation relies on a large number of pixel-level labels. However, it is difficult to obtain them. To this end, we present a method of Semi-supervised Infrared Ship Target Segmentation with Dual Branch (SeISTS-DB), which utilizes a small amount of labeled data and a large amount of unlabeled data to train model and improve segmentation performance. There are three main contributions. First, we design a target segmentation branch to generate the pseudo labels for unlabeled data. It consists of a dual learning network and a segmentation network. The dual learning network generates pseudo labels with weights for unlabeled data. The segmentation network is trained using both labeled data and unlabeled data with pseudo labels to achieve target segmentation of infrared ship, obtaining the preliminary segmentation results. Secondly, we introduce an error segmentation pixel correction branch, which contains a student network and a teacher network, to modify the pixel category error of the preliminary segmentation map. Finally, the outputs of the two branches are combined to obtain the final segmentation result. The SeISTS-DB is compared with other fully-supervised and semi-supervised methods on the infrared ship images dataset. Experimental results demonstrate that when the labeled data accounts for 1/8 of the training data, the mean Intersection over Union (mIou) is respectively improved by 15.35% and 6.19% at most. Besides, it is also compared with other methods on the public IRSTD-1k dataset, when the proportion of labeled images is 1/8, the mIoU is respectively improved by 11.76% at most compared to the state-of-the-art semi-supervised methods, demonstrating its effectiveness.

A Segmentation Framework with Unsupervised Learning-Based Label Mapper for the Ventricular Target of Intracranial Germ Cell Tumor.

Intracranial germ cell tumors are rare tumors that mainly affect children and adolescents. Radiotherapy is the cornerstone of interdisciplinary treatment methods. Radiation of the whole ventricle system and the local tumor can reduce the complications in the late stage of radiotherapy while ensuring the curative effect. However, manually delineating the ventricular system is labor-intensive and time-consuming for physicians. The diverse ventricle shape and the hydrocephalus-induced ventricle dilation increase the difficulty of automatic segmentation algorithms. Therefore, this study proposed a fully automatic segmentation framework. Firstly, we designed a novel unsupervised learning-based label mapper, which is used to handle the ventricle shape variations and obtain the preliminary segmentation result. Then, to boost the segmentation performance of the framework, we improved the region growth algorithm and combined the fully connected conditional random field to optimize the preliminary results from both regional and voxel scales. In the case of only one set of annotated data is required, the average time cost is 153.01s, and the average target segmentation accuracy can reach 84.69%. Furthermore, we verified the algorithm in practical clinical applications. The results demonstrate that our proposed method is beneficial for physicians to delineate radiotherapy targets, which is feasible and clinically practical, and may fill the gap of automatic delineation methods for the ventricular target of intracranial germ celltumors.

Automatic segmentation of thyroid with the assistance of the devised boundary improvement based on multicomponent small dataset.

Deep learning has been widely considered in medical image segmentation. However, the difficulty of acquiring medical images and labels can affect the accuracy of the segmentation results for deep learning methods. In this paper, an automatic segmentation method is proposed by devising a multicomponent neighborhood extreme learning machine to improve the boundary attention region of the preliminary segmentation results. The neighborhood features are acquired by training U-Nets with the multicomponent small dataset, which consists of original thyroid ultrasound images, Sobel edge images and superpixel images. Afterward, the neighborhood features are selected by min-redundancy and max-relevance filter in the designed extreme learning machine, and the selected features are used to train the extreme learning machine to obtain supplementary segmentation results. Finally, the accuracy of the segmentation results is improved by adjusting the boundary attention region of the preliminary segmentation results with the supplementary segmentation results. This method combines the advantages of deep learning and traditional machine learning, boosting the accuracy of thyroid segmentation accuracy with a small dataset in a multigroup test.

ESA-UNet for assisted diagnosis of cardiac magnetic resonance image based on the semantic segmentation of the heart.

Cardiovascular diseases have become the number one disease affecting human health in today's society. In the diagnosis of cardiac diseases, magnetic resonance image (MRI) technology is the most widely used one. However, in clinical diagnosis, the analysis of MRI relies on manual work, which is laborious and time-consuming, and also easily influenced by the subjective experience of doctors. In this article, we propose an artificial intelligence-aided diagnosis system for cardiac MRI with image segmentation as the main component to assist in the diagnosis of cardiovascular diseases. We first performed adequate pre-processing of MRI. The pre-processing steps include the detection of regions of interest of cardiac MRI data, as well as data normalization and data enhancement, and then we input the images after data pre-processing into the deep learning network module of ESA-Unet for the identification of the aorta in order to obtain preliminary segmentation results, and finally, the boundaries of the segmentation results are further optimized using conditional random fields. For ROI detection, we first use standard deviation filters for filtering to find regions in the heart cycle image sequence where pixel intensity varies strongly with time and then use Canny edge detection and Hough transform techniques to find the region of interest containing the heart. The ESA-Unet proposed in this article, moreover, is jointly designed with a self-attentive mechanism and multi-scale jump connection based on convolutional networks. The experimental dataset used in this article is from the Department of CT/MRI at the Second Affiliated Hospital of Fujian Medical University. Experiments compare other convolution-based methods, such as UNet, FCN, FPN, and PSPNet, and the results show that our model achieves the best results on Acc, Pr, ReCall, DSC, and IoU metrics. After comparative analysis, the experimental results show that the ESA-UNet network segmentation model designed in this article has higher accuracy, intuitiveness, and more application value than traditional image segmentation algorithms. With the continuous application of nuclear magnetic resonance technology in clinical diagnosis, the method in this article is expected to become a tool that can effectively improve the efficiency of doctors' diagnoses.

Real-Time Ship Segmentation in Maritime Surveillance Videos Using Automatically Annotated Synthetic Datasets.

This work proposes a new system capable of real-time ship instance segmentation during maritime surveillance missions by unmanned aerial vehicles using an onboard standard RGB camera. The implementation requires two stages: an instance segmentation network able to produce fast and reliable preliminary segmentation results and a post-processing 3D fully connected Conditional Random Field, which significantly improves segmentation results by exploring temporal correlations between nearby frames in video sequences. Moreover, due to the absence of maritime datasets consisting of properly labeled video sequences, we create a new dataset comprising synthetic video sequences of maritime surveillance scenarios (MarSyn). The main advantages of this approach are the possibility of generating a vast set of images and videos, being able to represent real-world scenarios without the necessity of deploying the real vehicle, and automatic labels, which eliminate human labeling errors. We train the system with the MarSyn dataset and with aerial footage from publicly available annotated maritime datasets to validate the proposed approach. We present some experimental results and compare them to other approaches, and we also illustrate the temporal stability provided by the second stage in missing frames and wrong segmentation scenarios.

Real time detection of inter-row ryegrass in wheat farms using deep learning

A key challenge for autonomous precision weeding is to reliably and accurately detect weed plants and crop plants in real time to minimise damage to surrounding crop plants while performing weeding actions. Specifically for a wheat farm, classifying ryegrass weed plants is particularly difficult even with human eyes since ryegrass shows visually very similar shape and texture to the crop plants themselves. A Deep Neural Network (DNN) that exploits the geometric location of ryegrass is proposed for the real time segmentation of inter-row ryegrass weeds in a wheat field. Our proposed method introduces two subnets in a conventional encoder-decoder style DNN to improve segmentation accuracy. The two subnets treat inter-row and intra-row pixels differently, and provide corrections to preliminary segmentation results of the conventional encoder-decoder DNN. A dataset captured in a wheat farm by an agricultural robot at different time instances is used to evaluate the segmentation performance, and the proposed method performs the best among various popular semantic segmentation algorithms. The proposed method runs at 48.95 Frames Per Second (FPS) with a consumer level graphics processing unit, thus is real-time deployable at camera frame rate.

3D Brain Tumor Image Segmentation Integrating Cascaded Anisotropic Fully Convolutional Neural Network and Hybrid Level Set Method

Abstract The accuracy of three-dimensional (3D) brain tumor image segmentation is of great significance to brain tumor diagnosis. To enhance the accuracy of segmentation, this study proposes an algorithm integrating a cascaded anisotropic fully convolutional neural network (FCNN) and the hybrid level set method. The algorithm first performs bias field correction and gray value normalization on T1, T1C, T2, and fluid-attenuated inversion recovery magnetic resonance imaging (MRI) images for preprocessing. It then uses a cascading mechanism to perform preliminary segmentation of whole tumors, tumor cores, and enhancing tumors by an anisotropic FCNN based on the relationships among the locations of the three types of tumor structures. This simplifies multiclass brain tumor image segmentation problems into three binary classification problems. At the same time, the anisotropic FCNN adopts dense connections and multiscale feature merging to further enhance performance. Model training is respectively conducted on the axial, coronal, and sagittal planes, and the segmentation results from the three different orthogonal views are combined. Finally, the hybrid level set method is adopted to refine the brain tumor boundaries in the preliminary segmentation results, thereby completing fine segmentation. The results indicate that the proposed algorithm can achieve 3D MRI brain tumor image segmentation of high accuracy and stability. Comparison of the whole-tumor, tumor-core, and enhancing-tumor segmentation results with the gold standards produced Dice similarity coefficients (Dice) of 0.9113, 0.8581, and 0.7976, respectively.

Real-time segmentation of remote sensing images with a combination of clustering and Bayesian approaches

In the area of remote sensing image processing, accurate segmentation of high-resolution remote sensing images in real time remains a challenging problem and numerous approaches have been developed for the problem. This paper proposes a new unsupervised approach that can efficiently analyze a remote sensing image and provide accurate segmentation results. The approach performs segmentation in three stages. In the first stage, an image is partitioned into blocks of equal sizes. The mean values of the R, G and B components of the pixels in each block are computed to form a feature vector of the block. A preliminary segmentation result is obtained by clustering the feature vectors with a simple clustering algorithm. In the second stage, a Bayesian approach is applied to refine the preliminary segmentation result. In the final stage, a graph-based method is utilized to recognize regions with complex texture structures. The performance of this approach has been tested on a few benchmark datasets, and its segmentation accuracy is compared with that of many state-of-the-art segmentation tools for remote sensing images. The testing results show that the overall segmentation accuracy of the proposed approach is higher than that of the other tools, and real-time analysis suggests that the approach is promising for real-time applications. An implementation of the approach in MATLAB is freely available upon request.

Using color and 3D geometry features to segment fruit point cloud and improve fruit recognition accuracy

Fruit image segmentation is an essential step to distinguish fruits from the background. In order to improve the fruits recognition accuracy for harvesting robots in three-dimensional (3D) space, a method with the fusion of color and 3D geometry features for fruit point cloud segmentation was proposed in this study. The local descriptor was applied to obtain the candidate regions, and the global descriptor was used to obtain the final segmented results. Firstly, the hue, saturation, value (HSV) color features and normal orientation features of pixels were fused to obtain the preliminary segmentation results. Then, the pre-processed color image and depth image were converted to a point cloud, and it was clustered into multiple regions by the Euclidean clustering algorithm. Finally, we utilized the viewpoint feature histogram (VFH) of each point cloud cluster to remove the remaining non-fruit regions. The experiments showed that the segmentation accuracy of the proposed method was 98.99%, and the precision was 80.09%, which are both superior to the traditional color segmentation methods. In addition, a fruit detection method based on shape analysis showed that it is more effective in improving fruit recognition rate and reducing false detection rate than the color segmentation methods.

RESEARCH ON IMPROVED REGION GROWING POINT CLOUD ALGORITHM

Abstract. The effective segmentation of point clouds is a prerequisite for surface reconstruction, blind spot repair, and so on. Among them, regional growth is widely used due to its simple and easy to implement algorithm. However, the traditional regional growth segmentation algorithm often causes problems such as over-segmentation or voiding of the segmentation result due to the instability of the local features of the point cloud or the unreasonable selection of the initial seed nodes. In view of the above shortcomings, this paper proposes an improved region growing point cloud algorithm. Firstly, by calculating the Gaussian curvature and the average curvature of the point cloud data and sorting them, and setting the minimum curvature point as the seed node, the total number of clusters is reduced, and the quality of the classification result is improved. Secondly, the growth of the point cloud region growth criterion is determined by combining the normal angles. Finally, according to the shape characteristics of the point cloud and the preliminary segmentation results, each threshold is adjusted and determined, and the segmentation result is optimized.The experimental results show that compared with the traditional regional growth segmentation algorithm, this method can not only reduce the total number of segmentation regions, but also segment the point cloud data quickly and effectively, and solve the segmentation result caused by the traditional region growth point cloud segmentation method. Problems such as stability improve the accuracy and stability of point cloud segmentation.

Pavement defect detection with fully convolutional network and an uncertainty framework

AbstractImage segmentation has been implemented for pavement defect detection, from which types, locations, and geometric information can be obtained. In this study, an integration of a fully convolutional network with a Gaussian‐conditional random field (G‐CRF), an uncertainty framework, and probability‐based rejection is proposed for detecting pavement defects. First, a fully convolutional network is designed to generate preliminary segmentation results, and a G‐CRF is used to refine the segmentation. Second, epistemic and aleatory uncertainties in the model and database are considered to overcome the disadvantages of traditional deep‐learning methods. Last, probability‐based rejection is conducted to remove unreasonable segmentations. The proposed method is evaluated on a data set of images that were obtained from 16 highways. The proposed integration segments pavement distresses from digital images with desirable performance. It also provides a satisfactory means to improve the accuracy and generalization performance of pavement defect detection without introducing a delay into the segmentation process.

Sample generation of semi‐automatic pavement crack labelling and robustness in detection of pavement diseases

Recent convolutional neural networks have made significant advancements in the detection of road cracks. However, the lack of accurate crack training data reduces the generalisation ability of the deep model. In this Letter, a semi-automatic pavement crack labelling algorithm is proposed to solve the problem of insufficient training data. First, the modified C-V model is used to obtain the preliminary segmentation results. Second, the direction of the initial segmentation area is calculated by the ellipse fitting method, and the preliminary segmentation results are used as samples for accurate labelling. Finally, a multi-scale feature extraction module is proposed for learning rich deep convolutional features, which allows the acquired crack features under a complex background to be more discriminant. The experimental results were compared with the manual marking method, and this method can achieve accurate marking of crack images with a low amount of interaction, thereby significantly reducing the cost of ground-truth making. The results of the validation and comparison experiments on test data sets indicate that the proposed method can not only effectively identify cracks, but also overcome the interference of many factors in the environment.

Fracture Segmentation Method Based on Contour Evolution and Gradient Direction Consistency in Sequence of Coal Rock CT Images

The coal rock exhibits obvious heterogeneity and anisotropy after a long-term geological evolution in nature. Therefore, the coal rock CT (computed tomography) image shows uneven grey scale, low contrast, and the fractures with weak boundaries. The accurate segmentation of the coal rock fracture networks is challenging. In this paper, a segmentation method of fractures based on contour evolution and gradient direction consistency is proposed to accurately segment the fracture networks in the sequence of coal rock CT images. According to the contour variation rule of the fractures in the discrete 3D (three dimensional) space formed by the sequence of CT images, the fracture contour evolution model (FCEM) is constructed and the preliminary segmentation results of fractures are obtained from FCEM. A 3D adaptive median filtering (3DAMF) and a 3D bilateral filtering (3DBF) are proposed. The high density miscellaneous point noises in the coal rock CT images are filtered by the 3DAMF. And the boundaries of fractures are enhanced by 3DBF. According to the similarity of the preliminary segmentation results of fractures and the real contours of fractures, the preliminary segmentation results of fractures are optimized based on the gradient direction consistency model (GDCM) proposed in this paper to obtain the accurate boundaries of fractures. The fracture segmentation method proposed in this paper can obtain accurate boundaries of fractures with weak boundaries, and the experimental results show that the segmentation efficiency for sequence is high and adaptability is strong.

Defect Detection of Aluminum Alloy Wheels in Radiography Images Using Adaptive Threshold and Morphological Reconstruction

In low-pressure casting, aluminum alloy wheels are prone to internal defects such as gas holes and shrinkage cavities, which call for X-ray inspection to ensure quality. Automatic defect segmentation of X-ray images is an important task in X-ray inspection of wheels. For this, a solution is proposed here that combines adaptive threshold segmentation algorithm and mathematical morphology reconstruction. First, the X-ray image of the wheel is smoothed, and then the smoothed image is subtracted from the original image, and the resulting difference image is binarized; the binary image resulting from the low threshold is taken as the marker image, and that from the high threshold is taken as mask image, and mathematical morphology reconstruction is performed on the two images, with the resulting image being the preliminary result of the wheel defect segmentation. Finally, with area and diameter parameters as the conditions, the preliminary segmentation result is analyzed, and the defect regions satisfying the conditions are taken as the ultimate result of the whole solution. Experiments proved the feasibility of the above solution, which is found capable of extracting different types of wheel defects satisfactorily.

A Cognitive Vision Method for Insect Pest Image Segmentation

Segmentation of pests from crop leaves is the prerequisite step for pest intelligent diagnosis. To improve the accuracy and stability of segmentation, a cognitive segmentation approach to pest images is present in this paper. The method works in the follow way. First, a pest image is divided into blocks via an image block processing method. Second, an adaptive learning algorithm is used to accurately select the initial cluster centers. Third, preliminary segmentation results are achieved using K-means clustering. Finally, three digital morphological features of an ellipse are adopted to remove leaf veins. Segmentation experiments were performed on crop images of whiteflies. Compared with the conventional methods, the proposed cognitive segmentation method was accurate and robust for segmentation of whitefly images, and will provide a foundation for further identifying these pests.

Automatic segmentation of the prostate on CT images using deep learning and multi-atlas fusion.

Automatic segmentation of the prostate on CT images has many applications in prostate cancer diagnosis and therapy. However, prostate CT image segmentation is challenging because of the low contrast of soft tissue on CT images. In this paper, we propose an automatic segmentation method by combining a deep learning method and multi-atlas refinement. First, instead of segmenting the whole image, we extract the region of interesting (ROI) to delete irrelevant regions. Then, we use the convolutional neural networks (CNN) to learn the deep features for distinguishing the prostate pixels from the non-prostate pixels in order to obtain the preliminary segmentation results. CNN can automatically learn the deep features adapting to the data, which are different from some handcrafted features. Finally, we select some similar atlases to refine the initial segmentation results. The proposed method has been evaluated on a dataset of 92 prostate CT images. Experimental results show that our method achieved a Dice similarity coefficient of 86.80% as compared to the manual segmentation. The deep learning based method can provide a useful tool for automatic segmentation of the prostate on CT images and thus can have a variety of clinical applications.

A Remote Sensing Industrial Solid Waste Image Segmentation Method Based on Improved Watershed Algorithm

Taking the high-resolution remote sensing image of industrial waste dump as the research object, this paper proposes a new image segmentation method based on markercontrolled watershed transform and region merging. The method gets the finial partition result by two-phrase segmentation on the pan-sharpened true color ALOS image. In the first phase, color gradient image of original image should be calculated and then it is modified by morphological impose minima, which should use markers extracted in two different ways. Lastly, preliminary segmentation result is obtained by watershed transform operates on the modified color gradient image. To solve the over-segmentation of industrial solid waste and other ground objects, region merging operation is performed according to the similarity measure criterion based on segmented objects’ color histogram Bhattacharyya coefficient in the second phrase, and then the final result is obtained. This method has been tested on the pan-sharpened ALOS image of 2.5 meters resolution in Shizuishan industrial zone, China. Experiment results demonstrate that this method is feasible and effective to segment the remote sensing industrial solid waste image.

Correlation-Based Discrimination Between Cardiac Tissue and Blood for Segmentation of the Left Ventricle in 3-D Echocardiographic Images

For automated segmentation of 3-D echocardiographic images, incorporation of temporal information may be helpful. In this study, optimal settings for calculation of temporal cross-correlations between subsequent time frames were determined, to obtain the maximum cross-correlation (MCC) values that provided the best contrast between blood and cardiac tissue over the entire cardiac cycle. Both contrast and boundary gradient quality measures were assessed to optimize MCC values with respect to signal choice (radiofrequency or envelope data) and axial window size. Optimal MCC values were incorporated into a deformable model to automatically segment the left ventricular cavity. MCC values were tested against, and combined with, filtered, demodulated radiofrequency data. Results reveal that using envelope data in combination with a relatively small axial window (0.7–1.25 mm) at fine scale results in optimal contrast and boundary gradient between the two tissues over the entire cardiac cycle. Preliminary segmentation results indicate that incorporation of MCC values has additional value for automated segmentation of the left ventricle.

Real-Time Tracking and Shape Analysis of Atrial Septal Defects in 3D Echocardiography

Real-time cardiac ultrasound (US) allows monitoring the heart motion during intracardiac beating heart procedures. Our application assists pediatric atrial septal defect (ASD) closure techniques using real-time 3D US guidance and rigid instruments. ASD tracking is also an important tool for facilitating systematic clinical studies of the dynamic behavior of the intra-atrial communication. One major image processing challenge is associated with the required processing of information at high frame rate, especially given the low image quality. We present an optimization scheme for a block flow technique, which combines the probability-based velocity computation for an entire block (a 3D volume centered on the ASD) with cyclic template matching. The adapted similarity imposes constraints both locally (from frame to frame) to conserve energy, and globally (from a reference template) to minimize cumulative errors. The algorithm is optimized for fast and reliable results. For tests, we use three intra-operational 4D ultrasound sequences of clinical infant beating hearts with ASD. Computing velocity at the block level with an optimized scheme, our technique tracks ASD motion at a frequency of 60 frames/s on clinical 4D datasets. Results are stable and accurate for changes in resolution and block size. In particular, we show robust real-time tracking and preliminary segmentation results of the ASD shape, size and orientation as a function of time. We present an optimized block flow technique for real-time tracking of ASD to assist in minimally invasive beating heart surgery. Our method proposes the standard use of references for processing repetitive data. This paper represents, to our knowledge, the first study on the dynamic morphology of ASD that takes into account the angular effect introduced by the slanted position of the intra-atrial communication with respect to the US probe.