Proceedings of the 2019 3rd International Conference on Advances in Image Processing最新文献

{"title":"A Survey of Hybrid Deep Learning Methods for Traffic Flow Prediction","authors":"Yan Shi, Haoran Feng, Xiongfei Geng, Xingui Tang, Yongcai Wang","doi":"10.1145/3373419.3373429","DOIUrl":"https://doi.org/10.1145/3373419.3373429","url":null,"abstract":"Traffic flow prediction using big data and deep learning attracts great attentions in recent years. Researchers show that DNN models can provide better traffic prediction accuracy than the traditional shallow models. Since the traffic flow reveals both spatial and temporal dependency characteristics, and may be impacted by weather, social event data etc., therefore, a set of hybrid DNN models have been presented recently in literature for further improving the traffic flow prediction performances. The hybrid models can capture dependency in multi-dimension and show better prediction performances than simple DNN models. This paper presents a thorough review and comparison of hybrid deep learning models for traffic flow prediction. We review the data sources used in hybrid deep learning and the various hybrid deep learning models built for trafficc flow prediction. The benefits of using hybrid models are summarized.","PeriodicalId":352528,"journal":{"name":"Proceedings of the 2019 3rd International Conference on Advances in Image Processing","volume":"40 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-11-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133853218","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}

{"title":"Effects of Color Systems' Transformation on Optical Flow Estimation of Noisy and Degraded Images","authors":"Syed Tafseer Haider Shah, Xuezhi Xiang","doi":"10.1145/3373419.3373457","DOIUrl":"https://doi.org/10.1145/3373419.3373457","url":null,"abstract":"Varying illumination and image blur are some of major challenges faced by contemporary methods of optical flow estimation. Despite significant advancement, these aspects have not received much of attention by modern-day methods. Latest work in this field is heavily affected and produce adverse results when dealing with images containing variable illumination and blur. In this paper, we investigate the effects of color space transformations on optical flow estimation from degraded and noisy images. In our experiments, clean and noisy images have been used. These images contain different kinds of blur and atmospheric effects such as fog, mist, shadows and dark regions. By estimating optical flow with three types of sequences in parallel (super clean, clean and noisy), and using four popular color systems, the effects of color space transformation have been observed on the estimated flow fields. The four color systems include RGB (red, blue, green), HSV (hue, saturation, value), HSL (hue, saturation, lightness) and XYZ (as standardized by the International Commission on Illumination in 1931). It is found that output of an optical flow algorithm not only depends on the color system adopted, but some color spaces tend to favor some special type of image sequences. For instance, XYZ color system is more favorable for the images abiding by the brightness constancy assumption while HSV color space is more suitable for blurry and noisy images. While keeping rest of the parameters unchanged but only transforming the color-space, we estimated the optical flow. Obviously the results of an algorithm applied to clean images for optical flow, would not be consistent with a flow estimated from same images containing noise. The objective is to compare this adversative effect for different color spaces. The flow estimation errors in four color systems have been reported and compared, and the best color-space is pointed out in each case. The paper also discusses the possible factors behind these variable outcomes with an insight into the basic frameworks of traditional methods for optical flow.","PeriodicalId":352528,"journal":{"name":"Proceedings of the 2019 3rd International Conference on Advances in Image Processing","volume":"11 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-11-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115274332","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}

{"title":"Is Satellite Image Target Camouflage Still Valid Under Deep Learning Target Detection?","authors":"Shuai Kang, Haichang Gao, Yiwen Tang, Yi Liu, Jiaming Chen","doi":"10.1145/3373419.3373448","DOIUrl":"https://doi.org/10.1145/3373419.3373448","url":null,"abstract":"Satellite images can be used to observe a wild range of ground in a bird's eye view. In the past, researchers used hand-craft features to detect targets in satellite images. With the rapid growth of deep learning, neural networks such as Faster R-CNN, YOLO, SSD and RetinaNet can detect targets precisely and quickly. Traditional camouflage is designed to make important targets hard to identify. So whether the satellite image target camouflage is still effective with deep learning target detection? In this paper, we use YOLO v3 and RetinaNet to verify the effectiveness of camouflage and propose an improved YOLO v3 to enhance detection efficiency, which raise the detection speed from 34.5fps to 55.3fps in an image of 416*416. The experimental results show that the target camouflage in the satellite image has no effect under deep learning target detection methods. At the end of the paper, suggestions on how to improve the camouflage effect to resist deep learning detection are proposed.","PeriodicalId":352528,"journal":{"name":"Proceedings of the 2019 3rd International Conference on Advances in Image Processing","volume":"144 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-11-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115310165","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}

{"title":"Research on Public Opinion of \"Chuang Youth Entrepreneurship Competition\" Based on Emotion Analysis","authors":"Chenyang Zhao, Peng Tian","doi":"10.1145/3373419.3373423","DOIUrl":"https://doi.org/10.1145/3373419.3373423","url":null,"abstract":"[purpose/significance] this paper takes weibo, a communication tool, as the carrier, and analyzes the emotional trend and emotional timing trend of netizens in the \"chuang youth entrepreneurial competition\" through weibo comments, so as to provide reference value for relevant departments in information communication and management and control of the event.[method/process] this paper takes weibo as the data source and \"chuang youth entrepreneurship competition\" as the theme keyword, collects the text information of netizens on chuang youth entrepreneurship competition, USES the emotion analysis method, constructs the public opinion time sequence chart, and analyzes the netizens' emotion situation on chuang youth entrepreneurship competition.[conclusion/result] the research results show that the public's attention and enthusiasm for entrepreneurship is not too high, only limited to specific groups of people to pay attention to this event.","PeriodicalId":352528,"journal":{"name":"Proceedings of the 2019 3rd International Conference on Advances in Image Processing","volume":"107 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-11-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124116858","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}

{"title":"Multispectral Images Pan-Sharpening Based on Atrous Convolution Network and Deep Residual Network","authors":"Tiantian Wang, Longshan Yang, Linlin Xu","doi":"10.1145/3373419.3373461","DOIUrl":"https://doi.org/10.1145/3373419.3373461","url":null,"abstract":"Pan-sharpening aims to fuse a panchromatic and a multispectral image to enhance the spatial resolution of the latter while retaining its spectral information. Although many algorithms for solving this task have been proposed, there is still room for improvement in spatial detail preservation. In this paper, we propose a network called ARNet to achieve multispectral image pan-sharpening through deep learning. In order to better preserve the spatial details in the multispectral image, we propose to obtain the prior information from the atrous convolution network and then combine it with the residual network (ResNet) to implement pan-sharpening. Experimental results of the quantitative and qualitative evaluation show that the proposed method outperforms state-of-the-art pan-sharpening methods.","PeriodicalId":352528,"journal":{"name":"Proceedings of the 2019 3rd International Conference on Advances in Image Processing","volume":"24 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-11-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121683562","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}

{"title":"Generating Pedestrian Training Dataset using DCGAN","authors":"Daeun Dana Kim, Muhammad Tanseef Shahid, Yunseong Kim, Won Jun Lee, H. Song, F. Piccialli, K. Choi","doi":"10.1145/3373419.3373458","DOIUrl":"https://doi.org/10.1145/3373419.3373458","url":null,"abstract":"Recently, as autonomous cars are developing very fast, it is the most crucial task to detect pedestrians for autonomous driving. Convolution neural network based on pedestrian detection models has gained enormous success in many applications. However, these models need a large amount of annotated and labeled datasets for training process which requires lots of time and human effort. For training samples, the diversity and quantity of datasets are very important. The proposed framework is based on Deep Convolutional Generative Adversarial Networks (DCGAN), able to generate realistic pedestrians. Experimental results show that DCGAN framework is able to synthesize real pedestrian images with diversity. The synthesized samples can be included in training data to improve the performance of pedestrian detectors. 24,770 images including PETA dataset, Inria dataset were used for the training process.","PeriodicalId":352528,"journal":{"name":"Proceedings of the 2019 3rd International Conference on Advances in Image Processing","volume":"128 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-11-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133854330","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}

{"title":"Design Big Data Analysis System - Bigdeepexaminator","authors":"J. Bobulski, M. Kubanek","doi":"10.1145/3373419.3373425","DOIUrl":"https://doi.org/10.1145/3373419.3373425","url":null,"abstract":"Big Data is a term used for such data sets, which at the same time are characterized by high volume, di-versity, real-time stream inflow, variability, complexity, as well as require the use of innovative technolo-gies, tools and methods in order to extracting new and useful knowledge from them. Big Data is a new challenge and information possibilities. Correct interpretation of data can play a key role in the global and local economy, social policy and enterprises. We present a data analysis system design with the use of ar-tificial intelligence that will help in obtaining valuable information from big data.","PeriodicalId":352528,"journal":{"name":"Proceedings of the 2019 3rd International Conference on Advances in Image Processing","volume":"168 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-11-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132656969","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}

{"title":"SAR Target Recognition Based on Cholesky Decomposition Weighted Kernel Extreme Learning Machine","authors":"Zijian Jin","doi":"10.1145/3373419.3373441","DOIUrl":"https://doi.org/10.1145/3373419.3373441","url":null,"abstract":"An SAR target recognition algorithm based on extreme learning machine is proposed. The traditional extreme learning machine cannot overcome the problem of sample noise and imbalance. To solve the problem, this paper introduces the weighted extreme learning machine algorithm, and uses KFCM algorithm combined with the proportion of samples to obtain the sample weight matrix. At the same time, in view of the problem that ordinary extreme learning machine uses matrix inverse to train the process, a calculation method based on choleksy decomposition is proposed. The experimental results show that the algorithm in this paper is faster and has higher recognition rate than ordinary algorithms such as KELM, SVM and BP neural network.","PeriodicalId":352528,"journal":{"name":"Proceedings of the 2019 3rd International Conference on Advances in Image Processing","volume":"2021 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-11-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123630436","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}

{"title":"An Ideology for the Prediction of Critical Haplotype Blocks of Variants in Genes (Cyp2c9 And Vkorc1) for Warfarin (Anticoagulant) Drug Dosage to Treat Heart Patients Efficiently by Using Ml (Machine Learning) and Data Stream Mining Techniques","authors":"Hina Saeeda, Muhammad Adil Abid","doi":"10.1145/3373419.3373424","DOIUrl":"https://doi.org/10.1145/3373419.3373424","url":null,"abstract":"Now a day's on time treatment of heart diseases is a very critical part of medical diagnoses. So far there are total 50 SNP (Single Nucleotide Polymorphism) diagnosed that are responsible for the heart problems. But it is very hard to study all of the SNP together because of their different base pairs' locations or changes in base pairs positions (variations in genetic code A C G T). These all 50 SNP are present in all individuals with different variations, it is a tough job to calculate all the changes in this SNP set as there are total of (50^50) positions to calculate which is making it a huge data set. For acquiring a data set of all these positions, we will need some good Data Stream Mining (data mining techniques) to find out all the possible locations of all the variants responsible for the heart problems. In this research paper, we are giving a short analysis and introduction to the problem of heart patients drug dosage associated with anticoagulant (Warfarin) and its risks, solution for the challenge of calculating all variants of two genes (CYP2C9 and VKORC1) and advantages of the proposed solution in the future.","PeriodicalId":352528,"journal":{"name":"Proceedings of the 2019 3rd International Conference on Advances in Image Processing","volume":"39 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-11-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122219686","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}

{"title":"An Image Visual Cryptography Using Double Encryption and Hiding Technology","authors":"Chen Pan, Xiao-ling Huang, G. Ye, Zhengxia Wang","doi":"10.1145/3373419.3373454","DOIUrl":"https://doi.org/10.1145/3373419.3373454","url":null,"abstract":"In this paper, an image visual cryptography based on double encryption and hiding technology is proposed. The secret plain-image is scrambled, added under modulo operation and cyclically left-shifted to obtain the pre-encrypted image. Carrier image is also encrypted by scrambling and diffusion, and then the pre-encrypted image is embedded into the encrypted carrier image by lifting wavelet transform (LWT) and QR decomposition. As a result, the final encrypted carrier image with encrypted secret image is generated. After realizing double visual encryption and hiding technology, attacker cannot know that the meaningless of encrypted carrier image containing an encrypted secret image. Chaotic sequences used in the scrambling and diffusion stages are generated by a 2D logistic-sine-coupling map (2D-LSCM), which is related to secret plain-image and can resist effectively the known-plaintext and chosen-plaintext attacks. Experimental results show that the proposed algorithm has high security and good application prospects.","PeriodicalId":352528,"journal":{"name":"Proceedings of the 2019 3rd International Conference on Advances in Image Processing","volume":"44 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-11-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127277836","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}