- 标题
- 摘要
- 关键词
- 实验方案
- 产品
-
FRED-Net: Fully Residual Encoder-Decoder Network for Accurate Iris Segmentation
摘要： Iris recognition is now developed enough to recognize a person from a distance. The process of iris segmentation plays a vital role in maintaining the accuracy of the iris-based recognition systems by limiting the errors at the current stage. However, its performance is affected by non-ideal situations created by environmental light noise and user non-cooperation. The existing local feature-based segmentation methods are unable to find the true iris boundary in these non-ideal situations, and the error created at the segmentation stage traverses to all the subsequent stages, which results in reduced accuracy and reliability. In addition, it is necessary to segment the true iris boundary without the extra cost of denoising as preprocessing. To overcome these challenging issues during iris segmentation, a deep learning-based fully residual encoder-decoder network (FRED-Net) is proposed to determine the true iris region with the flow of high-frequency information from the preceding layers via residual skip connection. The main four impacts and significances of this study are as follows. First, FRED-Net is an end-to-end semantic segmentation network that does not use conventional image processing schemes, and does not have a preprocessing overhead. It is a standalone network in which eyelid, eyelash, and glint detections are not required to obtain the true iris boundary. Second, the proposed FRED-Net is the final resultant structure of a step-by-step development, and in each step, a new complete variant network is created for semantic segmentation considering the detailed descriptions of the networks. Third, FRED-Net uses the residual connectivity between convolutional layers by the residual shortcut for both encoder and decoder, which enables a high-frequency component to flow through the network and achieve higher accuracy with few layers. Fourth, the performance of the proposed FRED-Net is tested with five different iris datasets under visible and NIR light environments, and two general road scene segmentation datasets. To achieve fair comparisons with other studies, our trained FRED-Net models, along with the algorithms, are made publicly available through our website (Dongguk FRED-Net Model with Algorithm. accessed on 16 May 2018). The experiments include two datasets: Noisy Iris Challenge Evaluation - Part II (NICE-II) selected from the UBIRIS.v2 database and Mobile Iris Challenge Evaluation (MICHE-I), for the visible light environment and three datasets: Institute of Automation, Chinese Academy of Sciences (CASIA) v4.0 interval, v4.0 distance, and IIT Delhi v1.0, for the near-infrared (NIR) light environment. Moreover, to evaluate the performance of the proposed network in general segmentation, experiments with two famous road scene segmentation datasets: Cambridge-driving Labeled Video Database (CamVid) and Karlsruhe Institute of Technology and Toyota Technological Institute at Chicago (KITTI), are included. The experimental results showed the optimum performance of the proposed FRED-Net on the above-mentioned seven datasets of iris and general road scene segmentation.
关键词： iris segmentation,full residual encoder-decoder network,Iris recognition,semantic segmentation
更新于2025-09-23 15:23:52
-
Preservation of image edge feature based on snowfall model smoothing filter
摘要： This paper proposed a snowfall model as a novel smoothing filter. The pixel composition of the image was similar to the geographic features, so it could be smooth because of snow accumulation. In the snowfall processing, luminance changes are linked to terrain and snowfall amount. Curvature and luminance gradient decided the amount of snowfall; the amount of snowfall became large on the parts where the curvature was large, and it became little on the parts where the gradient was steep. Snowfall algorithm simulates the natural snowfall process, which nonlinear diffusion and the target feature could be preserved well. Snowfall model has the same function as the Gaussian filter. The number of regions was reduced after Gaussian filter and snowfall model smoothing, respectively. The contrast experiment was carried out based on Watershed algorithm. The image area segmentation that pretreated through snowfall model was compared with Gaussian filter smoothing. The experimental result showed that the proposed snowfall model was a smoothing filter. It was able to realize edge preservation, which was the original purpose, and it was also possible to apply to region segmentation.
关键词： Snowfall model,Smoothing filter,Region segmentation,Edge characteristics,Image preservation
更新于2025-09-23 15:23:52
-
An Image Segmentation Method Based on Improved Regularized Level Set Model
摘要： When the level set algorithm is used to segment an image, the level set function must be initialized periodically to ensure that it remains a signed distance function (SDF). To avoid this defect, an improved regularized level set method-based image segmentation approach is presented. First, a new potential function is defined and introduced to reconstruct a new distance regularization term to solve this issue of periodically initializing the level set function. Second, by combining the distance regularization term with the internal and external energy terms, a new energy functional is developed. Then, the process of the new energy functional evolution is derived by using the calculus of variations and the steepest descent approach, and a partial differential equation is designed. Finally, an improved regularized level set-based image segmentation (IRLS-IS) method is proposed. Numerical experimental results demonstrate that the IRLS-IS method is not only effective and robust to segment noise and intensity-inhomogeneous images but can also analyze complex medical images well.
关键词： image segmentation,energy functional,level set,distance regularization term
更新于2025-09-23 15:23:52
-
Focal Boundary Guided Salient Object Detection
摘要： The performance of salient object segmentation has been significantly advanced by using deep convolutional networks. However, these networks often produce blob-like saliency maps without accurate object boundaries. This is caused by the limited spatial resolution of their feature maps after multiple pooling operations, and might hinder downstream applications that require precise object shapes. To address this issue, we propose a novel deep model—Focal Boundary Guided (Focal-BG) network. Our model is designed to jointly learn to segment salient object masks and detect salient object boundaries. Our key idea is that additional knowledge about object boundaries can help to precisely identify the shape of the object. Moreover, our model incorporates a refinement pathway to refine the mask prediction, and makes use of the focal loss to facilitate the learning of the hard boundary pixels. To evaluate our model, we conduct extensive experiments. Our Focal-BG network consistently outperforms state-of-the-art methods on five major benchmarks. We provide a detailed analysis of these results and demonstrate that our joint modeling of salient object boundary and mask helps to better capture shape details, especially in the vicinity of object boundaries.
关键词： Salient Object Segmentation,Deep Learning,Visual Saliency Detection,Boundary Detection
更新于2025-09-23 15:23:52
-
[IEEE 2018 Chinese Control And Decision Conference (CCDC) - Shenyang (2018.6.9-2018.6.11)] 2018 Chinese Control And Decision Conference (CCDC) - Optic disc segmentation method based on low rank matrix recovery theory
摘要： Optic disc(OD) detection and segmentation is one of the key elements for automatic retinal disease screening systems. This paper proposes an optic disc segmentation algorithm. The unified method based on low rank matrix recovery theory is used to make a significant detection. Then Hough transform is combined to obtain the final segmentation results. The proposed algorithm has short computation time and low complexity. It is robust to variable image illuminations and angles. The algorithm achieves the OD segmentation accuracy of 92.9% of the images on the MESSIDOR public dataset.
关键词： Optic disc,Hough transform,Segmentation,Low rank matrix recovery
更新于2025-09-23 15:23:52
-
RefMoB, a Reflectivity Feature Model-Based Automated Method for Measuring Four Outer Retinal Hyperreflective Bands in Optical Coherence Tomography
摘要： PURPOSE. To validate a model-driven method (RefMoB) of automatically describing the four outer retinal hyperreflective bands revealed by spectral-domain optical coherence tomography (SDOCT), for comparison with histology of normal macula; to report thickness and position of bands, particularly band 2 (ellipsoid zone [EZ], commonly called IS/OS). METHODS. Foveal and superior perifoveal scans of seven SDOCT volumes of five individuals aged 28 to 69 years with healthy maculas were used (seven eyes for validation, five eyes for measurement). RefMoB determines band thickness and position by a multistage procedure that models reflectivities as a summation of Gaussians. Band thickness and positions were compared with those obtained by manual evaluators for the same scans, and compared with an independent published histological dataset. RESULTS. Agreement among manual evaluators was moderate. Relative to manual evaluation, RefMoB reported reduced thickness and vertical shifts in band positions in a band-specific manner for both simulated and empirical data. In foveal and perifoveal scans, band 1 was thick relative to the anatomical external limiting membrane, band 2 aligned with the outer one-third of the anatomical IS ellipsoid, and band 3 (IZ, interdigitation of retinal pigment epithelium and photoreceptors) was cleanly delineated. CONCLUSIONS. RefMoB is suitable for automatic description of the location and thickness of the four outer retinal hyperreflective bands. Initial results suggest that band 2 aligns with the outer ellipsoid, thus supporting its recent designation as EZ. Automated and objective delineation of band 3 will help investigations of structural biomarkers of dark-adaptation changes in aging.
关键词： age-related macular degeneration,retina,ellipsoid,segmentation,optical coherence tomography,photoreceptors,interdigitation,reflectivity,retinal pigment epithelium
更新于2025-09-23 15:23:52
-
IRT image segmentation and enhancement using FCM-MALO approach
摘要： Infrared Thermography (IRT) is a method that has modernized the way for monitoring the thermal conditions, finding some potential faults or defects that could be available in electrical systems. In the proposed work, IRT electrical images are taken for diagnosing the faults by the image pre-processing and segmentation process. Initially, the IRT images are changed over into a grayscale image, trailed by image pre-processing is performed where histogram equalization is applied. With the intention of segmenting the faulty portion (high temperature zone) from the electrical equipment, Fuzzy C Means (FCM) strategy is introduced. For optimizing the centroid of FCM algorithm Modified Ant Lion Optimization (MALO) is proposed. From the segmented images, small size portions are removed by using Region Props function. This operation can remove the isolated pixels from the image and extract image components for better representation of images. The optimum results show that the proposed work accomplishes maximum segmentation accuracy compared to existing segmentation algorithms.
关键词： Pre-processing,Infrared thermography images,Fault diagnosis,Segmentation,Region props function,Electrical equipment
更新于2025-09-23 15:23:52
-
Segmentation-aided classification of hyperspectral data using spatial dependency of spectral bands
摘要： Classifying every pixel of a hyperspectral image with a certain land-cover type is the cornerstone of hyperspectral image analysis. In the present study a segmentation-aided methodology for the spectral-spatial classification of hyperspectral data is proposed. It considers the spatial dependence of the spectral bands, deals with the curse of dimensionality and handles the spectral variability. A local spatial regularization of spectral information is used, in order to derive an informative joint spectral-spatial representation of the data. A contiguity-based segmentation algorithm is formulated, in order to build the object-wise texture that can aid classifier learning. The hybrid use of the segmentation texture is evaluated in both pre-processing (i.e. selecting representative pixels to learn the classifier) and post-processing (i.e. refining predicted labels and removing possible outlier classifications). The experiments performed with the proposed methodology provide encouraging results, also compared to several recent state-of-the-art approaches.
关键词： Local spatial dependency analysis,Segmentation,Spectral-spatial classification,Curse of dimensionality
更新于2025-09-23 15:23:52
-
Scale-variable region-merging for high resolution remote sensing image segmentation
摘要： In high resolution remote sensing imagery (HRI), the sizes of different geo-objects often vary greatly, posing serious difficulties to their successful segmentation. Although existent segmentation approaches have provided some solutions to this problem, the complexity of HRI may still lead to great challenges for previous methods. In order to further enhance the quality of HRI segmentation, this paper proposes a new segmentation algorithm based on scale-variable region merging. Scale-variable means that the scale parameters (SP) adopted for segmentation are adaptively estimated, so that geo-objects of various sizes can be better segmented out. To implement the proposed technique, 3 steps are designed. The first step produces a coarse-segmentation result with slight degree of under segmentation error. This is achieved by segmenting a half size image with the global optimal SP. Such a SP is determined by using the image of original size. In the second step, structural and spatial contextual information is extracted from the coarse-segmentation, enabling the estimation of variable SPs. In the last step, a region merging process is initiated, and the SPs used to terminate this process are estimated based on the information obtained in the second step. The proposed method was tested by using 3 scenes of HRI with different landscape patterns. Experimental results indicated that our approach produced good segmentation accuracy, outperforming some competitive methods in comparison.
关键词： Image segmentation,High resolution remote sensing imagery,Scale-variable,Region merging
更新于2025-09-23 15:23:52
-
Biometric iris recognition using radial basis function neural network
摘要： The consistent and efficient method for the identification of biometrics is the iris recognition in view of the fact that it has richness in texture information. A good number of features performed in the past are built on handcrafted features. The proposed method is based on the feed-forward architecture and uses k-means clustering algorithm for the iris patterns classification. In this paper, segmentation of iris is performed using the circular Hough transform that realizes the iris boundaries in the eye and isolates the region of iris with no eyelashes and other constrictions. Moreover, Daugman's rubber sheet model is used to transform the resultant iris portion into polar coordinates in the process of normalization. A unique iris code is generated by log-Gabor filter to extract the features. The classification is achieved using neural network structures, the feed-forward neural network and the radial basis function neural network. The experiments have been conducted using the Chinese Academy of Sciences Institute of Automation (CASIA) iris database. The proposed system decreases computation time, size of the database and increases the recognition accuracy as compared to the existing algorithms.
关键词： Feed-forward neural network (FNN),Iris segmentation,Normalization,Biometrics,Radial basis function neural network (RBFNN),Iris recognition
更新于2025-09-23 15:23:52