Segment Boundaries Research Articles

Global video object segmentation with spatial constraint module

We present a lightweight and efficient semi-supervised video object segmentation network based on the space-time memory framework. To some extent, our method solves the two difficulties encountered in traditional video object segmentation: one is that the single frame calculation time is too long, and the other is that the current frame’s segmentation should use more information from past frames. The algorithm uses a global context (GC) module to achieve high-performance, real-time segmentation. The GC module can effectively integrate multi-frame image information without increased memory and can process each frame in real time. Moreover, the prediction mask of the previous frame is helpful for the segmentation of the current frame, so we input it into a spatial constraint module (SCM), which constrains the areas of segments in the current frame. The SCM effectively alleviates mismatching of similar targets yet consumes few additional resources. We added a refinement module to the decoder to improve boundary segmentation. Our model achieves state-of-the-art results on various datasets, scoring 80.1% on YouTube-VOS 2018 and a {cal J}{rm{& }}{cal F} score of 78.0% on DAVIS 2017, while taking 0.05 s per frame on the DAVIS 2016 validation dataset.

GUDU: Geometrically-constrained Ultrasound Data augmentation in U-Net for echocardiography semantic segmentation

Echocardiography is a very important medical examination that helps in the computation of critical heart functions. Boundary identification, segmentation and estimation of the volume of key parts of the heart, especially the left ventricle, is a difficult and time-consuming process, even for the most experienced cardiologists. In recent years, research has focused on the automatic segmentation of heart through artificial intelligence techniques and especially with the use of deep learning. Our work is part of this framework.We implemented an ensemble of convolutional neural networks based on the U-net architecture, trained it using a public dataset of cardiac ultrasound images, and combined the outcomes to extract the areas of the left ventricle, myocardium and left atrium. In order to optimize the training process, we have developed a significant data augmentation method based on medical practice. Furthermore, we extended the Dice loss function by imposing additional mandatory anatomical constraints. An ablation study highlights the contribution of each of our proposed modules.The evaluation of our method showed an overall improvement in segmentation accuracy but also in the estimation of clinical metrics. Specifically, using the Dice coefficient for geometric metrics, we achieved for the epicardium a score of 0.96 and 0.955 for the end-diastolic and end-systolic phase respectively. For the clinical metrics of the left ventricle volume, the Pearson correlation coefficient was used where our method gave 0.977, 0.981, 0.897 for the end-diastolic, end-systolic phase and ejection fraction respectively. Scores which up until the writing of this article outperform competitive methods.

MDE-UNet: A Multitask Deformable UNet Combined Enhancement Network for Farmland Boundary Segmentation

Farmland segmentation scenario from remote sensing images plays an important role in crop growth monitoring, precision agriculture and intelligent agriculture. To achieve high precision segmentation of farmland boundary, a Multi-task Deformable UNet combined Enhanced network (MDE-UNet) is proposed for farmland boundary segmentation. The network consists of two parts: a Multi-task Deformable UNet (MD-UNet) segmentation module with Deformable UNet (D-UNet) as the basic network and an enhancement module with a lightweight UNet improved by residual attention. In the MD-UNet segmentation module, three branches are used for precise segmentation of deterministic, fuzzy, and raw boundary, respectively. In the enhancement module, an improved lightweight UNet is designed, which can enhance the feature extraction ability of the MD-UNet segmentation module and further improve the segmentation accuracy. The accuracy and mIoU in the GF-2 farmland segmentation test dataset can reach 96.41% and 91.29% using the proposed model, respectively. The MDE-UNet method outperforms other representative deep learning methods such as DeepLab v3+, FCN-8s, SegFormer, and UTNet, and has potential for practical applications of farmland boundary segmentation.

Adaptive Boundary and Semantic Composite Segmentation Method for Individual Objects in Aerial Images

There are two types of methods for image segmentation. One is traditional image processing methods, which are sensitive to details and boundaries, yet fail to recognize semantic information. The other is deep learning methods, which can locate and identify different objects, but boundary identifications are not accurate enough. Both of them cannot generate entire segmentation information. In order to obtain accurate edge detection and semantic information, an Adaptive Boundary and Semantic Composite Segmentation method (ABSCS) is proposed. This method can precisely semantic segment individual objects in large-size aerial images with limited GPU performances. It includes adaptively dividing and modifying the aerial images with the proposed principles and methods, using the deep learning method to semantic segment and preprocess the small divided pieces, using three traditional methods to segment and preprocess original-size aerial images, adaptively selecting traditional results to modify the boundaries of individual objects in deep learning results, and combining the results of different objects. Individual object semantic segmentation experiments are conducted by using the AeroScapes dataset, and their results are analyzed qualitatively and quantitatively. The experimental results demonstrate that the proposed method can achieve more promising object boundaries than the original deep learning method. This work also demonstrates the advantages of the proposed method in applications of point cloud semantic segmentation and image inpainting.

Timely and Accurate Bitrate Switching in HTTP Adaptive Streaming With Date-Driven I-Frame Prediction

In today's Internet, bandwidth dynamics are inevitable, and hence, the bitrate for live streaming applications should also be dynamically adjusted. However, in existing HTTP-based adaptive streaming (HAS), bitrate switching can only be performed at segment boundaries, making decisions unresponsive and often inaccurate. In this paper, we start from a close investigation on the impact of the segment length in HAS and accordingly present VHAS, an extension towards intelligent variable-length segmentation, which makes client-side decisions based on the massive amount of real-time information from the network and viewers. VHAS implements a smart trigger mechanism that balances accuracy and overhead for variable-length segmentation. We further develop an adaptive bitrate switching algorithm with data-driven I-frame prediction, which is tailored to individual viewers to minimize bitrate mismatches. We evaluate VHAS via extensive trace-driven simulations, and our results demonstrate that compared with state-of-the-art solutions, VHAS achieves 15% - 49% gains in QoE, with a noticeable bandwidth reduction of 37% - 57%.

MIGN: Multiscale Image Generation Network for Remote Sensing Image Semantic Segmentation

With the development of computer vision, the semantic segmentation of remote sensing images, which has become an important topic, has been utilized in various applications for image content analysis and understanding, such as urban planning, natural disaster monitoring, and land resource management. Many approaches have been proposed to address these problems. However, due to obvious differences in resolution, spatial structure, and semantics between remote sensing images and ordinary images, the semantic segmentation of remote sensing images is still challenging. In this paper, we propose a novel multiscale image generation network (MIGN) that can efficiently generate high-resolution segmentation results by considering both details and boundary information. In particular, a multi-attention mechanism method for semantic segmentation of remote sensing images is designed. The attention weight is calculated by capturing the interaction of cross dimensions in a two-branch structure, which can learn the underlying feature information and guarantee the performance of each pixel feature for final classification. We also propose an edge supervised module to ensure that the segmentation boundary has a more accurate performance. A multiscale image fusion algorithm based on the Bayes model is proposed to improve the accuracy of the segmentation module. The performance of our model is evaluated on the ISPRS Vaihingen and Potsdam datasets. The results show that our method is superior to the most advanced image segmentation methods in terms of MIoU and pixel accuracy.

Rethinking and Improving Few-Shot Segmentation From a Contour-Aware Perspective

Existing few-shot segmentation approaches basically adopt the idea of comparing the semantic prototype vector of the query image and support images, and then obtaining the segmentation result. However, recent studies have shown that a single feature vector in feature map cannot accurately represent pixel-level categories, thus leading to poor segmentation of object boundary and semantic ambiguity. To address this common problem, we propose a novel contour-aware network (CTANet) for few-shot segmentation in this paper. Unlike the usual practice of classifying each pixel separately, CTANet regards all pixels within the same contour as a whole, which can take advantage of the internal consistency of objects to obtain a more accurate representation of category information. To obtain the accurate object contour, our network consists of a contour generation module and a contour refinement module, where the former exploits multiple levels of features to generate a primary contour map and the latter learns to refine the primary contour map. Furthermore, a novel contour-aware mixed loss is proposed to fuse the common BCE loss and our contour-aware loss to supervise the training process on two levels, pixel-level and contour-level. Extensive experiments demonstrate that our CTANet achieves a new state-of-the-art performance on <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$ \text{PASCAL-5}^{i}$</tex-math></inline-formula> and <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$ \text{COCO-20}^{i}$</tex-math></inline-formula> . Hopefully, our new perspective could provide more clues for future research on few-shot segmentation. Our code is freely available at: <uri xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">https://github.com/hardtogetA/CTANet</uri> .

Multiscale Attention Feature Aggregation Network for Cloud and Cloud Shadow Segmentation

Cloud and cloud shadow segmentation is one of the most important problems in remote sensing image processing. Due to the vulnerability to ground object interference, noise interference and other factors, and the lack of generalization ability, traditional deep learning network would inevitably lose details and spatial information, resulting in imprecise segmentation of cloud and cloud shadow boundaries, missing detection and false detection. In order to solve these problems, a multi-scale attention feature aggregation network is proposed, which extracts semantic information at different levels based on residual network. A multi-scale strip pooling attention module is designed to extract multi-scale context information and deep spatial channel information. In order to improve the feature extraction of the model, the deep multi-head feedforward transfer attention module is introduced to enable the adjacent two layers of the backbone network to guide each other for feature mining. Then, a bilateral feature fusion module is used to guide the fusion of low-level semantic information and high-level detail information. Finally, in order to enhance the repair of boundary details of cloud and cloud shadow, a boundary refinement boosting model is designed. The experimental results show that our model can handle complex cloud cover scenes, and has excellent performance on cloud, cloud shadow dataset and CSWV dataset. Its segmentation accuracy is superior to the existing methods, which is of great significance to the related work of cloud detection.

Optimizing Spatial Relationships in GCN to Improve the Classification Accuracy of Remote Sensing Images

Semantic segmentation of remote sensing images is one of the core tasks of remote sensing image interpretation. With the continuous development of artificial intelligence technology, the use of deep learning methods for interpreting remote-sensing images has matured. Existing neural networks disregard the spatial relationship between two targets in remote sensing images. Semantic segmentation models that combine convolutional neural networks (CNNs) and graph convolutional neural networks (GCNs) cause a lack of feature boundaries, which leads to the unsatisfactory segmentation of various target feature boundaries. In this paper, we propose a new semantic segmentation model for remote sensing images (called DGCN hereinafter), which combines deep semantic segmentation networks (DSSN) and GCNs. In the GCN module, a loss function for boundary information is employed to optimize the learning of spatial relationship features between the target features and their relationships. A hierarchical fusion method is utilized for feature fusion and classification to optimize the spatial relationship information in the original feature information. Extensive experiments on ISPRS 2D and DeepGlobe semantic segmentation datasets show that compared with the existing semantic segmentation models of remote sensing images, the DGCN significantly optimizes the segmentation effect of feature boundaries, effectively reduces the noise in the segmentation results and improves the segmentation accuracy, which demonstrates the advancements of our model.

О методах адаптивной сегментации при квазистатическом анализе многопроводных линий передачи

Various methods of adaptive segmentation of the cross-section boundaries of multiconductor transmission lines are generalized. The method of moments was used in the simulation. Four methods of adaptive segmentation are considered and the most optimal of them is determined. The optimality criterion was accuracy and cost-effectiveness of calculating matrices of per-unit-length parameters of transmission lines. For verification, 6 different software tools and measurement results were used.

Instance Motion Tendency Learning for Video Panoptic Segmentation.

Video panoptic segmentation is an important but challenging task in computer vision. It not only performs panoptic segmentation of each frame, but also associates the same instance across adjacent frames. Due to the lack of temporal coherence modeling, most existing approaches often generate identity switches during instance association, and they cannot handle ambiguous segmentation boundaries caused by motion blur. To address these difficult issues, we introduce a simple yet effective Instance Motion Tendency Network (IMTNet) for video panoptic segmentation. It learns a global motion tendency map for instance association, and a hierarchical classifier for motion boundary refinement. Specifically, a Global Motion Tendency Module (GMTM) is designed to learn robust motion features from optical flows, which can directly associate each instance in the previous frame to the corresponding instance in the current frame. In addition, we propose a Motion Boundary Refinement Module (MBRM) to learn a hierarchical classifier to handle the boundary pixels of moving targets, which can effectively revise the inaccurate segmentation predictions. Experimental results on both Cityscapes and Cityscapes-VPS datasets show that our IMTNet outperforms most state-of-the-art approaches.

Bootstrapping Semi-supervised Medical Image Segmentation with Anatomical-Aware Contrastive Distillation.

Contrastive learning has shown great promise over annotation scarcity problems in the context of medical image segmentation. Existing approaches typically assume a balanced class distribution for both labeled and unlabeled medical images. However, medical image data in reality is commonly imbalanced (i.e., multi-class label imbalance), which naturally yields blurry contours and usually incorrectly labels rare objects. Moreover, it remains unclear whether all negative samples are equally negative. In this work, we present ACTION, an Anatomical-aware ConTrastive dIstillatiON framework, for semi-supervised medical image segmentation. Specifically, we first develop an iterative contrastive distillation algorithm by softly labeling the negatives rather than binary supervision between positive and negative pairs. We also capture more semantically similar features from the randomly chosen negative set compared to the positives to enforce the diversity of the sampled data. Second, we raise a more important question: Can we really handle imbalanced samples to yield better performance? Hence, the key innovation in ACTION is to learn global semantic relationship across the entire dataset and local anatomical features among the neighbouring pixels with minimal additional memory footprint. During the training, we introduce anatomical contrast by actively sampling a sparse set of hard negative pixels, which can generate smoother segmentation boundaries and more accurate predictions. Extensive experiments across two benchmark datasets and different unlabeled settings show that ACTION significantly outperforms the current state-of-the-art semi-supervised methods.

The temporal evolution of seismicity and variability of b-values along the Vienna Basin Transfer Fault System

Abstract The Vienna Basin Transfer Fault System (VBTFS) is the most active fault system in the region between the Eastern Alps, the western Carpathians and the Pannonian Basin. The spatial and temporal distribution of earthquakes along the fault system shows a heterogeneous pattern including a long-time decay of seismicity at the northern part of the VBTFS, which was interpreted to result from a long aftershock sequence subsequent to the 1906 Dobrá Voda earthquake (M=5.7). In this paper we investigate if other segments of the VBTFS display similar long-term declines of seismicity that might indicate long aftershock sequences following strong, yet unrecorded, earthquakes in historical times. In order to analyse the distribution of seismicity, the VBTFS is divided into arbitrary segments of about 50 km length each. The segments are chosen to overlap each other to avoid missing information from neighbouring segments due to arbitrarily selected segment boundaries. For each segment we analyse the temporal evolution of seismicity and calculate the parameters of the corresponding Gutenberg-Richter (GR) relation. The temporal seismicity patterns revealed from the segments covering the Dobrá Voda area confirm the protracted aftershock sequence following the 1906 earthquake. All but one of the other segments do not show temporal changes of seismicity comparable to the long-term Dobrá Voda aftershock sequence. Seismicity patterns, however, include short-term Omori-type aftershocks following moderate earthquakes such as the 2000 Ebreichsdorf earthquake (M=4.8). The segment covering the SW tip of the VBTFS revealed a 200 years long gradual decrease of the largest observed magnitudes starting with the 1794 Leoben (M=4.7) earthquake. The 1794 event is the oldest earthquake listed in the catalogue for the region under consideration. It therefore remains open if the recorded decay of seismicity results from the 1794 event, or a stronger earthquake before that time. The latter is corroborated by the low magnitude of the 1794 earthquake which would typically not be considered to cause long aftershock sequences. GR a- and b-values, calculated for the individual segments, vary significantly along the VBTFS. Values range from 0.47 to 0.86 (b-values) and 0.81 to 2.54 (a-values), respectively. Data show a significant positive correlation of a- and b-values and a coincidence of the lowest b-values with fault segments with large seismic slip deficits and very low seismicity in the last approximately 300 years. These parts of the VBTFS were previously interpreted as “locked” fault segments, which have a significant potential to release future strong earthquakes, in spite of the fact that historical and instrumentally recorded seismicity is very low. We find this interpretation corroborated by the low b-values that suggest high differential stresses for these fault segments.

A CIF-Based Speech Segmentation Method for Streaming E2E ASR

Long utterances segmentation is crucial in end-to-end (E2E) streaming automatic speech recognition (ASR). However, commonly used voice activity detection(VAD)-based and fixed-length segmentation methods may lead to long segments and semantic incompleteness, affecting the user experience and ASR performance. In this paper, we propose a speech segmentation method for streaming E2E ASR to solve the above issues. Both the decoder's dependence on acoustic information and the human average breath frequency are used for judging segment boundaries. Frame-level decoder's dependence information is provided by the Continuous Integrate-and-Fire (CIF) predictor, which optimizes jointly with ASR to guarantee a more suitable segmentation for ASR. Besides, the proposed method does not increase the model parameters and real-time factor (RTF). The experimental results show that our method can accurately detect the pauses in speech, and the segment usually contains relatively complete semantic information. Compared with VAD-based segmentation, 53.5% latency reduction and 3.7% CER reduction relatively are achieved.

Підвищення стійкості цифрових водяних знаків у потокових відеозаписах на основі диференціального вбудовування енергії (DEW)

In connection with the continuous growth of the amount of digital information and the wider use of the Internet, today’s urgent task is the protection of copyright with the help of steganography. In particular, the technologies of digital watermarks (DWM) are successfully used to solve the problems of confirming copyright in relation to digital files, controlling their use by authorized users and other persons, confirming the authenticity and reliability of copies of software and multimedia objects used. An algorithm for embedding digital watermarks based on the idea of a differential energy watermark (DEW) was chosen for research and further improvement in the work, and is a method of embedding a digital watermark based on the selective rejection of a part of the high-frequency coefficients of the discrete cosine transform (DCT) compressed images and videos. Based on the results of the work carried out, this article describes the improvement of the method for copyright protection of streaming video (MPEG2 and MPEG4 formats) by embedding a digital watermark only in the low-frequency coefficients of the discrete-cosine transformation and applying the Canni’s segmentation method, which uses a multi-level computational approach to the detection of the segment boundaries in the image and helps to reduce the visual changes of the container. In order to study the practical results of the improved method, it was implemented in the developed online service and applied to video files, which made it possible to evaluate the performance and compare it with other methods of embedding the DW in terms of visual impact indicators (SSIM, SNR, PSNR, AD, NAD, IF) and resistance to attacks. According to the results of the study, it was concluded that in general, the application of the improved method allows to increase the stability indicators of DW in streaming video by approximately 12%. The digital watermark embedded in the video file did not have a significant impact on it, and the embedding of data in low-frequency coefficients made it possible to increase resistance to attacks.

CdcSegNet: Automatic COVID-19 Infection Segmentation From CT Images

It has been more than two years since the outbreak of COVID-19, which has spread to almost every corner of the world and killed a great number of people. Rapid detection and screening have become an important means of controlling the spread of COVID-19. Segmentation of COVID-19 infected tissue from computed tomography (CT) images of a patient’s lungs can provide clinicians with important information to quantify and diagnose COVID-19. However, the accuracy of medical image segmentation is seriously affected by such factors as the low contrast between the infected tissue and the edge of the surrounding environment, the large variation of the infected tissue and the lack of labeling data. Therefore, a deep learning model called CdcSegNet to accurately segment lung lesions from CT images infected by COVID-19 is proposed. In our method, transfer learning is introduced to solve the problem of lack of annotation data, and three modules, i.e., continuous dilated convolution module (CDC), parallel dual attention module (PDA) and additional multi-core pooling layer (AMP) are innovatively proposed to solve the problem of fuzzy segmentation boundary and to segment effectively infected tissues. Extensive experiments and comparison studies are made, and demonstrate that our model CdcSegNet has high accuracy in COVID-19 segmentation, and is superior to the state-of-the-art models in terms of DICE, SEN, SPE, PPV, and VOE.

Construction of Improved Semantic Segmentation Model and Application to Extraction of Anthropogenically Disturbed Parcels With Soil Erosion From Remote Sensing Images

With the rapid socioeconomic development in China, the increasing soil erosion caused by anthropogenic production and construction activities is taking place, which is characterized by short duration, high frequency, and great damage to its surrounding environment. Therefore, the regulation and control of soil erosion of anthropogenically disturbed parcels is an urgent task. This study proposes an improved model that combines the boundary constraint and jagged hybrid dilated convolution channel shuffling module (BCJHDC) and the polarized self-attention (PSA) module for extracting anthropogenically disturbed parcels with soil erosion (ADPSE) from high-resolution remote sensing images in Hubei Province. Firstly, the PSA module is added to the encoder to better extract the feature information of the target object. Secondly, BCJHDC module is used to extract multi-scale semantic information from images and improve the boundary segmentation quality. Precision, recall, intersection over union (IOU), and F1 score are calculated to evaluate the model accuracy. The results indicate that our improved model performs well on the human-perturbed parcel extraction task with an F1 of 87.92% and an IOU of 78.44%. Ablation experiments and application experiments suggest the validity the applicability, and the portability of our proposed improved model, respectively. Compared with the other 7 advanced semantic segmentation models, our improved model has significant advantages. Overall, this study provides valuable reference for policy formulation of water and soil conservation.

Contour-Aware Equipotential Learning for Semantic Segmentation

With increasing demands for high-quality semantic segmentation in the industry, hard-distinguishing semantic boundaries have posed a significant threat to existing solutions. Inspired by real-life experience, i.e., combining varied observations contributes to higher visual recognition confidence, we present the equipotential learning (EPL) method. This novel module transfers the predicted/ground-truth semantic labels to a self-defined potential domain to learn and infer decision boundaries along customized directions. The conversion to the potential domain is implemented via a lightweight differentiable anisotropic convolution that neither incurs the parameter overhead of the neural networks nor changes to the architectures. Besides, the designed two loss functions, the point loss and the equipotential line loss implement anisotropic field regression and category-level contour learning, respectively, enhancing prediction consistencies in the inter/intra-class boundary areas. More importantly, EPL is agnostic to network architectures, and thus it can be plugged into most existing segmentation models. This paper is the first attempt to address the boundary segmentation problem with field regression and contour learning. Meaningful performance improvements on Pascal Voc 2012 and Cityscapes demonstrate that the proposed EPL module can benefit the off-the-shelf fully convolutional network models when recognizing semantic boundary areas. Besides, intensive comparisons and analysis show the favorable merits of EPL for distinguishing semantically-similar and irregular-shaped categories.

Segmentation boundaries in accelerometer data of arm motion induced by music: Online computation and perceptual assessment

Segmentation is a cognitive process involved in the understanding of information perceived through the senses. Likewise, the automatic segmentation of data captured by sensors may be used for the identification of patterns. This study is concerned with the segmentation of dancing motion captured by accelerometry and its possible applications, such as pattern learning and recognition, or gestural control of devices. To that effect, an automatic segmentation system was formulated and tested. Two participants were asked to ‘dance with one arm’ while their motion was measured by an accelerometer. The performances were recorded on video, and manually segmented by six annotators later. The annotations were used to optimize the automatic segmentation system, maximizing a novel similarity score between computed and annotated segmentations. The computed segmentations with highest similarity to each annotation were then manually assessed by the annotators, resulting in Precision between 0.71 and 0.89, and Recall between 0.82 to 1.

Lens structure segmentation from AS-OCT images via shape-based learning

Background and objectivesThe lens is one of the important refractive media in the eyeball. Abnormality of the nucleus or cortex in the lens can lead to ocular disorders such as cataracts and presbyopia. To achieve an accurate diagnosis, segmentation of these ocular structures from anterior segment optical coherence tomography (AS-OCT) is essential. However, weak-contrast boundaries of the object in the images present a challenge for accurate segmentation. The state-of-the-art (SOTA) methods, such as U-Net, treat segmentation as a binary classification of pixels, which cannot handle pixels on weak-contrast boundaries well. MethodsIn this paper, we propose to incorporate shape prior into a deep learning framework for accurate nucleus and cortex segmentation. Specifically, we propose to learn a level set function, whose zero-level set represents the object boundary, through a convolutional neural network. Moreover, we design a novel shape-based loss function, where the shape prior knowledge can be naturally embedded into the learning procedure, leading to improvement in performance. We collect a high-quality AS-OCT image dataset with precise annotations to train our model. ResultsAbundant experiments are conducted to verify the effectiveness of the proposed framework and the novel shape-based loss. The mean Intersection over Unions (MIoUs) of the proposed method for lens nucleus and cortex segmentation are 0.946 and 0.957, and the mean Euclidean Distance (MED) measure, which can reflect the accuracy of the segmentation boundary, are 6.746 and 2.045 pixels. In addition, the proposed shape-based loss improves the SOTA models on the nucleus and cortex segmentation tasks by an average of 0.0156 and 0.0078 in the MIoU metric and 1.394 and 0.134 pixels in the MED metric. ConclusionWe transform the segmentation from a classification task to a regression task by making the model learn the level set function, and embed shape information in deep learning by designing loss functions. This allows the proposed method to be more efficient in the segmentation of the object with weak-contrast boundaries. Concise abstractWe propose to incorporate shape priors into a deep learning framework for accurate nucleus and cortex segmentation from AS-OCT images. Specifically, we propose to learn a level set function, where the zero-level set represents the boundary of the target. Meanwhile, we design a novel shape-based loss function in which additional convex shape prior can be embedded in the learning process, leading to an improvement in performance. The IOUs for nucleus and cortex segmentation are 0.946 and 0.957, while the MED that reflects the accuracy of the boundary are 6.746 and 2.045 pixels. The proposed shape-based loss improves the SOTA model for nucleus and cortex segmentation by an average of 0.0156 and 0.0078 in IOU, and 1.394 and 0.134 pixels in MED. We transform segmentation from classification to regression by making the model learn a level set function, resulting in improved performance at the boundary with weak contrast.