Pub Date : 2018-12-01DOI: 10.1109/DICTA.2018.8615830
Md.Atiqur Rahman, S. Islam, Jungpil Shin, Md. Rashedul Islam
The intention of data compression is to promote the storage and delivery of big images with excellent compression ratio and least distortion. Moreover, the number of internet user is growing day by day speedily. Therefore, the transfer of data is being another significant concern. The storage and the use of an uncompressed picture are very costly and time-consuming. There are many techniques such as Arithmetic coding, Run-length coding, Huffman coding, Shannon-Fano coding used to compress an image. Compression of a picture using the state-of-the-art techniques has a high impact. However, The compression ratio and transfer speed do not satisfy the current demand. This article proposes a new histogram alternation based lossy image compression using Base-2 coding. It increases the probabilities of an image by doing a little bit of change to its pixels level which helps to reduce code-word. This algorithm uses less storage space and works at high-speed to encode and decode an image. Average code length, compression ratio, mean square error and pick signal to noise ratio are used to estimate this method. The proposed method demonstrates better performance than the state-of-the-art techniques.
{"title":"Histogram Alternation Based Digital Image Compression using Base-2 Coding","authors":"Md.Atiqur Rahman, S. Islam, Jungpil Shin, Md. Rashedul Islam","doi":"10.1109/DICTA.2018.8615830","DOIUrl":"https://doi.org/10.1109/DICTA.2018.8615830","url":null,"abstract":"The intention of data compression is to promote the storage and delivery of big images with excellent compression ratio and least distortion. Moreover, the number of internet user is growing day by day speedily. Therefore, the transfer of data is being another significant concern. The storage and the use of an uncompressed picture are very costly and time-consuming. There are many techniques such as Arithmetic coding, Run-length coding, Huffman coding, Shannon-Fano coding used to compress an image. Compression of a picture using the state-of-the-art techniques has a high impact. However, The compression ratio and transfer speed do not satisfy the current demand. This article proposes a new histogram alternation based lossy image compression using Base-2 coding. It increases the probabilities of an image by doing a little bit of change to its pixels level which helps to reduce code-word. This algorithm uses less storage space and works at high-speed to encode and decode an image. Average code length, compression ratio, mean square error and pick signal to noise ratio are used to estimate this method. The proposed method demonstrates better performance than the state-of-the-art techniques.","PeriodicalId":130057,"journal":{"name":"2018 Digital Image Computing: Techniques and Applications (DICTA)","volume":"19 1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116099329","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2018-12-01DOI: 10.1109/DICTA.2018.8615859
Yongliang Chen, W. Gong, Chaoyue Chen, Weihong Li
Benefiting from the great success of deep learning in computer vision, object detection with Convolutional Neural Network (CNN) based methods have drawn significant attentions. Various frameworks have been proposed which show awesome and robust performance for a large range of datasets. However, for building detection in remote sensing images, buildings always pose a diversity of orientation which makes it a challenge for the application of off-the-shelf methods to building detection in remote sensing images. In this work, we aim to integrate orientation regression into the popular axis-aligned bounding box to tackle this problem. To adapt the axis-aligned bounding boxes to arbitrarily orientated ones, we also develop an algorithm to estimate the Intersection Over Union (IOU) overlap between any two arbitrarily oriented boxes which is convenient to implement in Graphics Processing Unit (GPU) for fast computation. The proposed method utilizes CNN for both robust feature extraction and bounding box regression. We present our model in an end-to-end fashion making it easy to train. The model is formulated and trained to predict both orientation and location simultaneously obtaining tighter bounding box and hence, higher mean average precision (mAP). Experiments on remote sensing images of different scales shows a promising performance over the conventional one.
得益于深度学习在计算机视觉领域的巨大成功，基于卷积神经网络(CNN)的目标检测方法备受关注。已经提出了各种框架，它们在大范围的数据集上表现出令人敬畏和强大的性能。然而，对于遥感图像中的建筑物检测来说，建筑物总是具有多样性的方向，这给现成的方法在遥感图像中建筑物检测中的应用带来了挑战。在这项工作中，我们的目标是将方向回归集成到流行的轴对齐边界框中来解决这个问题。为了使轴线对齐的边界框适应于任意方向的边界框，我们还开发了一种算法来估计任意两个任意方向的边界框之间的交联(Intersection Over Union, IOU)重叠，该算法便于在图形处理单元(GPU)中实现，以实现快速计算。该方法利用CNN进行鲁棒特征提取和边界盒回归。我们以端到端方式呈现我们的模型，使其易于训练。模型的制定和训练可以同时预测方向和位置，从而获得更紧密的边界框，从而获得更高的平均精度(mAP)。在不同尺度遥感图像上的实验表明，该方法优于传统方法。
{"title":"Learning Orientation-Estimation Convolutional Neural Network for Building Detection in Optical Remote Sensing Image","authors":"Yongliang Chen, W. Gong, Chaoyue Chen, Weihong Li","doi":"10.1109/DICTA.2018.8615859","DOIUrl":"https://doi.org/10.1109/DICTA.2018.8615859","url":null,"abstract":"Benefiting from the great success of deep learning in computer vision, object detection with Convolutional Neural Network (CNN) based methods have drawn significant attentions. Various frameworks have been proposed which show awesome and robust performance for a large range of datasets. However, for building detection in remote sensing images, buildings always pose a diversity of orientation which makes it a challenge for the application of off-the-shelf methods to building detection in remote sensing images. In this work, we aim to integrate orientation regression into the popular axis-aligned bounding box to tackle this problem. To adapt the axis-aligned bounding boxes to arbitrarily orientated ones, we also develop an algorithm to estimate the Intersection Over Union (IOU) overlap between any two arbitrarily oriented boxes which is convenient to implement in Graphics Processing Unit (GPU) for fast computation. The proposed method utilizes CNN for both robust feature extraction and bounding box regression. We present our model in an end-to-end fashion making it easy to train. The model is formulated and trained to predict both orientation and location simultaneously obtaining tighter bounding box and hence, higher mean average precision (mAP). Experiments on remote sensing images of different scales shows a promising performance over the conventional one.","PeriodicalId":130057,"journal":{"name":"2018 Digital Image Computing: Techniques and Applications (DICTA)","volume":"41 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122493453","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2018-12-01DOI: 10.1109/DICTA.2018.8615845
Asghar Ali, M. Pickering
In this paper, a new Convolutional Neural Network (CNN) architecture is proposed for synthetic Urdu and English character recognition in natural scene images. The features are extracted using three separate sub-models of the CNN which are then fused in one feature vector. The network is purely trained on the synthetic character images of English and Urdu texts in natural images. For English text, the Chars74k-Font dataset is used and for Urdu text, the synthetic dataset is created by automatically cropping the image patches from four background image datasets and then putting characters at random positions within the image patch. The network is evaluated on a combined synthetic dataset of English and Urdu characters and the separate synthetic characters of Urdu and English datasets. The experimental results show that the network performs well on synthetic datasets.
{"title":"Feature-Level Fusion using Convolutional Neural Network for Multi-Language Synthetic Character Recognition in Natual Images","authors":"Asghar Ali, M. Pickering","doi":"10.1109/DICTA.2018.8615845","DOIUrl":"https://doi.org/10.1109/DICTA.2018.8615845","url":null,"abstract":"In this paper, a new Convolutional Neural Network (CNN) architecture is proposed for synthetic Urdu and English character recognition in natural scene images. The features are extracted using three separate sub-models of the CNN which are then fused in one feature vector. The network is purely trained on the synthetic character images of English and Urdu texts in natural images. For English text, the Chars74k-Font dataset is used and for Urdu text, the synthetic dataset is created by automatically cropping the image patches from four background image datasets and then putting characters at random positions within the image patch. The network is evaluated on a combined synthetic dataset of English and Urdu characters and the separate synthetic characters of Urdu and English datasets. The experimental results show that the network performs well on synthetic datasets.","PeriodicalId":130057,"journal":{"name":"2018 Digital Image Computing: Techniques and Applications (DICTA)","volume":"61 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122950408","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2018-12-01DOI: 10.1109/dicta.2018.8615755
{"title":"List of Meta Reviewers","authors":"","doi":"10.1109/dicta.2018.8615755","DOIUrl":"https://doi.org/10.1109/dicta.2018.8615755","url":null,"abstract":"","PeriodicalId":130057,"journal":{"name":"2018 Digital Image Computing: Techniques and Applications (DICTA)","volume":"43 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124940074","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2018-12-01DOI: 10.1109/DICTA.2018.8615821
Yue Li, R. Mathew, D. Taubman
Light fields, as a type of visual content, richer in textural and geometric information than traditional imaging, can exhibit strong redundancies between views. Disparity compensated prediction, as one of the view synthesis frameworks, can exploit these redundancies to achieve high coding efficiency. Properly handling texture occlusion in the prediction process is important. We propose a disparity guided texture inpainting scheme to resolve texture occlusion. It turns out that reliable disparity (depth) can be available within occluded regions. A key contribution of this paper is the incorporation of disparity to guide the pixel visiting order and the weighted-average interpolation processes of the inpainting scheme. Specifically, the paper describes a disparity-dependent boundary distance metric, which is evaluated using a Dijkstra's algorithm and used to drive inpainting decisions. Our proposed method is evaluated on a realistic dataset with complex geometry, presenting promising results.
{"title":"Disparity Guided Texture Inpainting for Light Field View Synthesis","authors":"Yue Li, R. Mathew, D. Taubman","doi":"10.1109/DICTA.2018.8615821","DOIUrl":"https://doi.org/10.1109/DICTA.2018.8615821","url":null,"abstract":"Light fields, as a type of visual content, richer in textural and geometric information than traditional imaging, can exhibit strong redundancies between views. Disparity compensated prediction, as one of the view synthesis frameworks, can exploit these redundancies to achieve high coding efficiency. Properly handling texture occlusion in the prediction process is important. We propose a disparity guided texture inpainting scheme to resolve texture occlusion. It turns out that reliable disparity (depth) can be available within occluded regions. A key contribution of this paper is the incorporation of disparity to guide the pixel visiting order and the weighted-average interpolation processes of the inpainting scheme. Specifically, the paper describes a disparity-dependent boundary distance metric, which is evaluated using a Dijkstra's algorithm and used to drive inpainting decisions. Our proposed method is evaluated on a realistic dataset with complex geometry, presenting promising results.","PeriodicalId":130057,"journal":{"name":"2018 Digital Image Computing: Techniques and Applications (DICTA)","volume":"19 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126167562","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2018-12-01DOI: 10.1109/DICTA.2018.8615834
K. Islam, S. Wijewickrema, Masud Pervez, S. O'Leary
Natural road trail image classification is a challenging problem due to the complexity of the natural road environment. It is useful in many real-world applications such as autonomous vehicle and robot navigation. In recent years, many researchers have explored the use of data obtained from different sensors in solving this problem. In this paper, we use image data captured from standard digital cameras, to address the road trail classification problem. To this end, we develop a database of road trail images and train an artificial neural network (ANN) classifier on features obtained using the bag-of-words (BoW) image feature extraction approach. We show experimentally that the proposed method is effective in classifying road trails.
{"title":"Road Trail Classification using Color Images for Autonomous Vehicle Navigation","authors":"K. Islam, S. Wijewickrema, Masud Pervez, S. O'Leary","doi":"10.1109/DICTA.2018.8615834","DOIUrl":"https://doi.org/10.1109/DICTA.2018.8615834","url":null,"abstract":"Natural road trail image classification is a challenging problem due to the complexity of the natural road environment. It is useful in many real-world applications such as autonomous vehicle and robot navigation. In recent years, many researchers have explored the use of data obtained from different sensors in solving this problem. In this paper, we use image data captured from standard digital cameras, to address the road trail classification problem. To this end, we develop a database of road trail images and train an artificial neural network (ANN) classifier on features obtained using the bag-of-words (BoW) image feature extraction approach. We show experimentally that the proposed method is effective in classifying road trails.","PeriodicalId":130057,"journal":{"name":"2018 Digital Image Computing: Techniques and Applications (DICTA)","volume":"23 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125223062","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2018-12-01DOI: 10.1109/DICTA.2018.8615839
Rangwan Kasantikul, Worapan Kusakunniran
This paper proposes the novel microaneurysm segmentation technique, based on the autoencoder-regularized neural network model. The proposed method is developed using two levels of the segmentation. First, the coarse-level segmentation stage locates the candidate areas using the multi-scale correlation filter and region growing. Second, the fine-level segmentation stage uses the neural network to obtain confidence values of candidate areas of being microaneurysm. The neural network based technique introduced in this paper is the modified multilayer neural network with an additional branch to take into account of the reconstruction error (in a similar fashion to the autoencoder). This modification to the neural network results in the consistent improvement in the classification performance, when compared to the conventional network without such modification. The proposed method is evaluated using the retinopathic online challenge dataset. It can deliver very promising results, when compared with the existing state-of-the-art techniques.
{"title":"Improving Supervised Microaneurysm Segmentation using Autoencoder-Regularized Neural Network","authors":"Rangwan Kasantikul, Worapan Kusakunniran","doi":"10.1109/DICTA.2018.8615839","DOIUrl":"https://doi.org/10.1109/DICTA.2018.8615839","url":null,"abstract":"This paper proposes the novel microaneurysm segmentation technique, based on the autoencoder-regularized neural network model. The proposed method is developed using two levels of the segmentation. First, the coarse-level segmentation stage locates the candidate areas using the multi-scale correlation filter and region growing. Second, the fine-level segmentation stage uses the neural network to obtain confidence values of candidate areas of being microaneurysm. The neural network based technique introduced in this paper is the modified multilayer neural network with an additional branch to take into account of the reconstruction error (in a similar fashion to the autoencoder). This modification to the neural network results in the consistent improvement in the classification performance, when compared to the conventional network without such modification. The proposed method is evaluated using the retinopathic online challenge dataset. It can deliver very promising results, when compared with the existing state-of-the-art techniques.","PeriodicalId":130057,"journal":{"name":"2018 Digital Image Computing: Techniques and Applications (DICTA)","volume":"92 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121757627","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2018-12-01DOI: 10.1109/DICTA.2018.8615873
Junpeng Zhang, X. Jia, Jiankun Hu, Kun Tan
Tracking multiple vehicles in optical satellite videos can be achieved by detecting vehicles in individual frames and then link detections across frames. This is challenging since the low spatial resolution of satellite videos would lead to miss or partial detections. Failure in consistently detecting an identical vehicle would terminate a trajectory for loss of identity, or reinitialize a new trajectory for re-detection. Both would result in fragmented tracklets for an identical vehicle. In this paper, we propose a two-step global data association approach for multiple target tracking, which first generates local tracklets then merges them to global trajectories. In order to handle the frequent terminals of tracklets caused by the miss or partial detections, the spatial grid flow model has been extended to temporal neighboring grids to cover the possible connectivities in a wider temporal neighboring, so that the detection association could match temporal-unlinked detections. In the tracklet association model, we customize a tracklet transition probability based on Kalman Filter to link tracklets with large temporal intervals. A near-optimal solution to the proposed two-step association problem is found by Bilevel K-shortest Paths Optimization. Experiments on a satellite video show that our approach improves both detection and tracking performance of moving vehicles.
{"title":"Satellite Multi-Vehicle Tracking under Inconsistent Detection Conditions by Bilevel K-Shortest Paths Optimization","authors":"Junpeng Zhang, X. Jia, Jiankun Hu, Kun Tan","doi":"10.1109/DICTA.2018.8615873","DOIUrl":"https://doi.org/10.1109/DICTA.2018.8615873","url":null,"abstract":"Tracking multiple vehicles in optical satellite videos can be achieved by detecting vehicles in individual frames and then link detections across frames. This is challenging since the low spatial resolution of satellite videos would lead to miss or partial detections. Failure in consistently detecting an identical vehicle would terminate a trajectory for loss of identity, or reinitialize a new trajectory for re-detection. Both would result in fragmented tracklets for an identical vehicle. In this paper, we propose a two-step global data association approach for multiple target tracking, which first generates local tracklets then merges them to global trajectories. In order to handle the frequent terminals of tracklets caused by the miss or partial detections, the spatial grid flow model has been extended to temporal neighboring grids to cover the possible connectivities in a wider temporal neighboring, so that the detection association could match temporal-unlinked detections. In the tracklet association model, we customize a tracklet transition probability based on Kalman Filter to link tracklets with large temporal intervals. A near-optimal solution to the proposed two-step association problem is found by Bilevel K-shortest Paths Optimization. Experiments on a satellite video show that our approach improves both detection and tracking performance of moving vehicles.","PeriodicalId":130057,"journal":{"name":"2018 Digital Image Computing: Techniques and Applications (DICTA)","volume":"8 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114357518","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2018-12-01DOI: 10.1109/DICTA.2018.8615848
Y. Shin, M. Sagong, Yoon-Jae Yeo, S. Ko
Image captured in the low-light environments usually suffers from the low dynamic ranges and noise which degrade the quality of the image. Recently, convolutional neural network (CNN) has been employed for low-light image enhancement to simultaneously perform the brightness enhancement and noise removal. Although conventional CNN based techniques exhibit superior performance compared to traditional non-CNN based methods, they often produce the image with visual artifacts due to the small receptive field in their network. In order to cope with this problem, we propose an adversarial context aggregation network (ACA-net) for low-light image enhancement, which effectively aggregates the global context via full-resolution intermediate layers. In the proposed method, we first increase the brightness of a low-light image using the two different gamma correction functions and then feed the brightened images to CNN to obtain the enhanced image. To this end, we train ACA network using L1 pixel-wise reconstruction loss and adversarial loss which encourages the network to generate a natural image. Experimental results show that the proposed method achieves state-of-the-art results in terms of peak signal-to-noise ratio (PSNR) and structural similarity index measure (SSIM).1
{"title":"Adversarial Context Aggregation Network for Low-Light Image Enhancement","authors":"Y. Shin, M. Sagong, Yoon-Jae Yeo, S. Ko","doi":"10.1109/DICTA.2018.8615848","DOIUrl":"https://doi.org/10.1109/DICTA.2018.8615848","url":null,"abstract":"Image captured in the low-light environments usually suffers from the low dynamic ranges and noise which degrade the quality of the image. Recently, convolutional neural network (CNN) has been employed for low-light image enhancement to simultaneously perform the brightness enhancement and noise removal. Although conventional CNN based techniques exhibit superior performance compared to traditional non-CNN based methods, they often produce the image with visual artifacts due to the small receptive field in their network. In order to cope with this problem, we propose an adversarial context aggregation network (ACA-net) for low-light image enhancement, which effectively aggregates the global context via full-resolution intermediate layers. In the proposed method, we first increase the brightness of a low-light image using the two different gamma correction functions and then feed the brightened images to CNN to obtain the enhanced image. To this end, we train ACA network using L1 pixel-wise reconstruction loss and adversarial loss which encourages the network to generate a natural image. Experimental results show that the proposed method achieves state-of-the-art results in terms of peak signal-to-noise ratio (PSNR) and structural similarity index measure (SSIM).1","PeriodicalId":130057,"journal":{"name":"2018 Digital Image Computing: Techniques and Applications (DICTA)","volume":"93 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124045596","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2018-12-01DOI: 10.1109/DICTA.2018.8615761
A. Garifullin, Peeter Koobi, Pasi Ylitepsa, Kati Adjers, M. Hauta-Kasari, H. Uusitalo, L. Lensu
The most common approach for retinal imaging is the eye fundus photography which usually results in RGB images. Recent studies show that the additional spectral information provides useful features for automatic retinal image analysis. The current work extends recent research on the joint segmentation of retinal vasculature, optic disc and macula which often appears in different retinal image analysis tasks. Fully convolutional neural networks are utilized to solve the segmentation problem. It is shown that the network architectures can be effectively modified for the spectral data and the utilization of spectral information provides moderate improvements in retinal image segmentation.
{"title":"Hyperspectral Image Segmentation of Retinal Vasculature, Optic Disc and Macula","authors":"A. Garifullin, Peeter Koobi, Pasi Ylitepsa, Kati Adjers, M. Hauta-Kasari, H. Uusitalo, L. Lensu","doi":"10.1109/DICTA.2018.8615761","DOIUrl":"https://doi.org/10.1109/DICTA.2018.8615761","url":null,"abstract":"The most common approach for retinal imaging is the eye fundus photography which usually results in RGB images. Recent studies show that the additional spectral information provides useful features for automatic retinal image analysis. The current work extends recent research on the joint segmentation of retinal vasculature, optic disc and macula which often appears in different retinal image analysis tasks. Fully convolutional neural networks are utilized to solve the segmentation problem. It is shown that the network architectures can be effectively modified for the spectral data and the utilization of spectral information provides moderate improvements in retinal image segmentation.","PeriodicalId":130057,"journal":{"name":"2018 Digital Image Computing: Techniques and Applications (DICTA)","volume":"15 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132048848","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：