- 标题
- 摘要
- 关键词
- 实验方案
- 产品
-
[IEEE 2018 International Joint Conference on Neural Networks (IJCNN) - Rio de Janeiro, Brazil (2018.7.8-2018.7.13)] 2018 International Joint Conference on Neural Networks (IJCNN) - Semantic Image Segmentation Based on Attentions to Intra Scales and Inner Channels
摘要： Multi-scale features provide different context information of objects, which is significant for better performance in semantic segmentation tasks. But different scale features contribute equally to final predicitons. In this paper, we propose a new attention mechanism that not only learns weights between different scales but also allocates importance to subregions of inner channels. The network architecture is built on a state-of-the-art feedforward network to generate strong semantic feature maps, accompanying with a top-down pathway that incorporates larger scale feature maps through lateral connections. The proposed intra-scale attention module softly weights features between scales pixel by pixel. To enhance impact of each feature map in intermediate layers on performance, we further present an inner-channel attention module to pay attention to subregions within each channel. Moreover, an extra supervision is presented to achieve excellent performance. Importantly, the inner-channel attention module adaptively changes features as the layer goes deeper and could be inserted into any other layers. Extensive experiments are conducted on PASCAL VOC2012 to verify the network effectiveness. The experimental results show that both intra-scale and inner-channel attention modules could yield better performance.
关键词： semantic segmentation,attention mechanism,inner-channel attention,intra-scale attention,multi-scale features
更新于2025-09-11 14:15:04
-
[IEEE 2018 International Seminar on Application for Technology of Information and Communication (iSemantic) - Semarang, Indonesia (2018.9.21-2018.9.22)] 2018 International Seminar on Application for Technology of Information and Communication - Improvement of Fuzzy C-Mean Using Local Laplacian Filter for Image Segmentation
摘要： the FCM (Fuzzy C-Mean) is on of algorithms used in the background image separation research. This study aims to improve the quality of image segmentation using FCM algorithm with Local Laplacian Filter. Thus, we applied the Local Laplacian Filter into the V (Value) and S (Saturation) component of the image. The result of Laplacian Filter then clustered using FCM method. The result of segmentation using FCM and Local Laplacian Filter will be compared with ground-truths images to reveal the value of the PSNR (Peak Signal to Noise Ratio) and the MSE (Mean Square Error). This study also compared the results of the MSE and PSNR with the FCM and K-Mean algorithms. The is done preprocessing first by using Local Laplacian Filter which gives the color contrast to the image. So the image becomes sharper and when the image changed to the color space HSV already looks quite striking color differences between objects with the background. The results of that segmentation using FCM and Local Laplacian Filter has the best MSE and PSNR results compared to the 2 tested algorithms.
关键词： HSV,Local Laplacian Filter,FCM,Image Segmentation
更新于2025-09-11 14:15:04
-
[IEEE 2017 IEEE International Conference on Computational Intelligence and Computing Research (ICCIC) - Coimbatore, India (2017.12.14-2017.12.16)] 2017 IEEE International Conference on Computational Intelligence and Computing Research (ICCIC) - A Survey on Advanced Segmentation Techniques in Image Processing Applications
摘要： Segmentation is considered as one of the main steps in image processing. To be simple, segmentation is nothing but partitioning an image. An image consists of foreground and background regions. Segmentation helps in separating these two regions for accurate analysis. Many techniques have been developed in recent years. These techniques are used to make the image look smoother for better analysis. This paper deals with the detailed survey on various image segmentation techniques involved in image processing applications.
关键词： region based segmentation,edge based segmentation,Image segmentation,color gradient,histogram image
更新于2025-09-10 09:29:36
-
Multiscale Optimized Segmentation of Urban Green Cover in High Resolution Remote Sensing Image
摘要： The urban green cover in high-spatial resolution (HR) remote sensing images have obvious multiscale characteristics, it is thus not possible to properly segment all features using a single segmentation scale because over-segmentation or under-segmentation often occurs. In this study, an unsupervised cross-scale optimization method specifically for urban green cover segmentation is proposed. A global optimal segmentation is first selected from multiscale segmentation results by using an optimization indicator. The regions in the global optimal segmentation are then isolated into under- and fine-segmentation parts. The under-segmentation regions are further locally refined by using the same indicator as that in global optimization. Finally, the fine-segmentation part and the refined under-segmentation part are combined to obtain the final cross-scale optimized result. The green cover objects can be segmented at their specific optimal segmentation scales in the optimized segmentation result to reduce both under- and over-segmentation errors. Experimental results on two test HR datasets verify the effectiveness of the proposed method.
关键词： scale parameter,multiscale segmentation,urban green cover,segmentation refinement,cross-scale optimization
更新于2025-09-10 09:29:36
-
An Improved Fuzzy Connectedness Method for Automatic Three-Dimensional Liver Vessel Segmentation in CT Images
摘要： In this paper, an improved fuzzy connectedness (FC) method was proposed for automatic three-dimensional (3D) liver vessel segmentation in computed tomography (CT) images. The vessel-enhanced image (i.e., vesselness image) was incorporated into the fuzzy a?nity function of FC, rather than the intensity image used by traditional FC. An improved vesselness ?lter was proposed by incorporating adaptive sigmoid ?ltering and a background-suppressing item. The fuzzy scene of FC was automatically initialized by using the Otsu segmentation algorithm and one single seed generated adaptively, while traditional FC required multiple seeds. The improved FC method was evaluated on 40 cases of clinical CT volumetric images from the 3Dircadb (n ? 20) and Sliver07 (n ? 20) datasets. Experimental results showed that the proposed liver vessel segmentation strategy could achieve better segmentation performance than traditional FC, region growing, and threshold level set. Average accuracy, sensitivity, speci?city, and Dice coe?cient of the improved FC method were, respectively, (96.4 ± 1.1)%, (73.7 ± 7.6)%, (97.4 ± 1.3)%, and (67.3 ± 5.7)% for the 3Dircadb dataset and (96.8 ± 0.6)%, (89.1 ± 6.8)%, (97.6 ± 1.1)%, and (71.4 ± 7.6)% for the Sliver07 dataset. It was concluded that the improved FC may be used as a new method for automatic 3D segmentation of liver vessel from CT images.
关键词： vesselness filter,fuzzy connectedness,CT images,liver vessel segmentation,Otsu segmentation algorithm
更新于2025-09-10 09:29:36
-
Fusion algorithm of infrared and visible images based on frame difference detection technology and area feature
摘要： A kind of fusion algorithm of infrared and visible images based on frame difference detection technology and area feature was proposed to increase the fusion quality of the infrared image and reduce complexity. Firstly, frame difference method was designed to complete detection on objectives in infrared images so as to conduct objective clustering and image segmentation; information among frames is used to complete accurate locating of objective; then different fusion rules were designed to complement effective information of visible light and infrared images as possible and complete image fusion according to features of objective area. The complexity of fusion algorithm in this thesis was analyzed theoretically. Meanwhile, fusion experiment was executed on two conditions of unmovable and observable objective in image of visible light and infrared light and movable and observable objective in image of visible light and infrared light; experiment result indicates that proposed technology has higher fusion quality and the fusion image can accurately reflect objective and background compared with current image fusion technology.
关键词： area feature,Image fusion,objective area segmentation,frame difference detection,infrared and visible light image
更新于2025-09-10 09:29:36
-
[IEEE 2018 25th IEEE International Conference on Image Processing (ICIP) - Athens, Greece (2018.10.7-2018.10.10)] 2018 25th IEEE International Conference on Image Processing (ICIP) - Joint Estimation of Local Variance and Local Regularity for Texture Segmentation. Application to Multiphase Flow Characterization
摘要： Texture segmentation constitutes a task of utmost importance in statistical image processing. Focusing on the broad class of monofractal textures characterized by piecewise constancy of the statistics of their multiscale representations, recently shown to be versatile enough for real-world texture modeling, the present work renews this recurrent topic by proposing an original approach enrolling jointly scale-free and local variance descriptors into a convex, but nonsmooth, minimization strategy. The performance of the proposed joint approach are compared against disjoint strategies working independently on scale-free features and on local variance on synthetic piecewise monofractal textures. Performance are also compared for multiphase flow image characterization, a topic of crucial importance in geophysics as well as in industrial processes. Applied to large-size images (above two million pixels), the proposed approach is shown to significantly improve state-of-the-art strategies by permitting the detection of the smallest gas bubbles and by offering a better understanding of multiphase flow structures.
关键词： Strong convexity,Primal-dual proximal algorithm,Multiphase flow,Texture segmentation,Total Variation
更新于2025-09-10 09:29:36
-
[IEEE IGARSS 2018 - 2018 IEEE International Geoscience and Remote Sensing Symposium - Valencia, Spain (2018.7.22-2018.7.27)] IGARSS 2018 - 2018 IEEE International Geoscience and Remote Sensing Symposium - Decision Fusion of Spot6 And Multitemporal Sentinel2 Images For Urban Area Detection
摘要： Fusion of very high spatial resolution multispectral (VHR) images and lower spatial resolution image time series with more spectral bands can improve land cover classification, combining geometric and semantic advantages of both sources. This study presents a workflow to extract the extent of urban areas using decision-level fusion of individual classifications on Sentinel 2 (S2) and SPOT6 satellite images. First, both sources are classified individually in five classes, using state-of-the-art supervised classification approaches and Convolutional Neural Networks. Obtained results are merged in order to extract buildings as accurately as possible. Then, detected buildings are merged again with the S2 classification to extract urban area; a prior to be in an urban area is derived from these building objects and merged with a binary classification derived from the original S2 classification. Both fusions involve a per pixel decision level fusion followed by a contrast sensitive regularization.
关键词： Regularization,Multispectral,Segmentation,Decision fusion,Urban classification,Urban area
更新于2025-09-10 09:29:36
-
[IEEE IGARSS 2018 - 2018 IEEE International Geoscience and Remote Sensing Symposium - Valencia, Spain (2018.7.22-2018.7.27)] IGARSS 2018 - 2018 IEEE International Geoscience and Remote Sensing Symposium - Joint Feature Network for Bridge Segmentation in Remote Sensing Images
摘要： This paper proposes a novel convolutional neural network architecture for semantic segmentation of bridges with various scales in optical remote sensing images. In the context of RSI analysis on objects with irregular shapes, it is necessary to get dense, pixelwise classification maps. To address the issue, a new network architecture for producing refined shapes is required instead of image categorization labels. In our end-to-end framework, a ResNet is used as a backbone model to extract semantic features, then a cascaded top-down path is added to fuse these features as different scales. Joint features are obtained by stacking different layers of feature maps. Experiments show our proposed architecture has the ability to combine rich multi-scale contextual information to produce semantic segmentation maps with high accuracy.
关键词： remote sensing images (RSIs),semantic segmentation,convolutional neural networks (CNNs),pixelwise classification
更新于2025-09-10 09:29:36
-
[IEEE 2018 40th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC) - Honolulu, HI, USA (2018.7.18-2018.7.21)] 2018 40th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC) - OCT Fluid Segmentation using Graph Shortest Path and Convolutional Neural Network <sup>*</sup>
摘要： Diagnosis and monitoring of retina diseases related to pathologies such as accumulated fluid can be performed using optical coherence tomography (OCT). OCT acquires a series of 2D slices (Bscans). This work presents a fully-automated method based on graph shortest path algorithms and convolutional neural network (CNN) to segment and detect three types of fluid including sub-retinal fluid (SRF), intra-retinal fluid (IRF) and pigment epithelium detachment (PED) in OCT Bscans of subjects with age-related macular degeneration (AMD) and retinal vein occlusion (RVO) or diabetic retinopathy. The proposed method achieves an average dice coefficient of 76.44%, 92.25% and 82.14% in Cirrus, Spectralis and Topcon datasets, respectively. The effectiveness of the proposed methods was also demonstrated in segmenting fluid in OCT images from the 2017 Retouch challenge.
关键词： optical coherence tomography,retinal vein occlusion,fluid segmentation,graph shortest path,convolutional neural network,age-related macular degeneration
更新于2025-09-10 09:29:36