The wide application of drones brings convenience to all aspects of society, but also seriously threatens the safety of the low-altitude field. In recent years, the number of accidents caused by UAVs is increasing, and it is urgent to study effective signal recognition technology for UAV targets in the low-altitude field. Therefore, this paper introduces the categories of UAV signals and explains the basic principles of signal recognition technology. Then it concludes a variety of signal recognition technologies and compares and analyzes the performance of existing technologies. Finally, it summarizes and prospects for the UAV signal recognition technology.
{"title":"UAV signal recognition technology","authors":"Yin Xue, Yuan-Lung Chang, Yu Zhang, Jiajun Ma, Guangjie Li, Q. Zhan, Dandan Wu, Jiancun Zuo","doi":"10.1145/3569966.3571186","DOIUrl":"https://doi.org/10.1145/3569966.3571186","url":null,"abstract":"The wide application of drones brings convenience to all aspects of society, but also seriously threatens the safety of the low-altitude field. In recent years, the number of accidents caused by UAVs is increasing, and it is urgent to study effective signal recognition technology for UAV targets in the low-altitude field. Therefore, this paper introduces the categories of UAV signals and explains the basic principles of signal recognition technology. Then it concludes a variety of signal recognition technologies and compares and analyzes the performance of existing technologies. Finally, it summarizes and prospects for the UAV signal recognition technology.","PeriodicalId":145580,"journal":{"name":"Proceedings of the 5th International Conference on Computer Science and Software Engineering","volume":"18 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-10-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117211072","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}
With the rapid development of computer vision, 3D data is increasing rapidly. How to retrieve similar model from a large number of models has become a hot research topic. However, in order to meet people's demand, the retrieval accuracy need to be further improved. In terms of multi-view 3D model retrieval, how to effectively learn the information between views is the key to improving performance. In this paper, we propose a novel 3D model retrieval algorithm based on attention and multi-view fusion. Specifically, we mainly constructed two modules. First, dynamic attentive graph learning module is used to learn the intrinsic relationship between view blocks; Then we propose the Attention-NetVlad algorithm, which combines the channel attention algorithm and the NetVlad algorithm. It learns the information between feature channels to enhance the feature expression ability firstly, then uses the NetVlad algorithm to fuse multiple view features into a global feature according to the clustering information. Finally the global feature is used as the only feature of the model to retrieve according to Euclidean distance. In comparison with other state-of-the-art methods by utilizing ModelNet10 and ModelNet40 the proposed method has demonstrated significant improvement for retrieval mAP. Our experiments also demonstrate the effectiveness of the modules in the algorithm.
{"title":"3D Model Retrieval Algorithm Based on Attention and Multi-view Fusion","authors":"Ziqi Shi, Ziyang Quan, Jingshan Shi, Zhuyan Guo, Mandun Zhang, Zhidong Xiao","doi":"10.1145/3569966.3570092","DOIUrl":"https://doi.org/10.1145/3569966.3570092","url":null,"abstract":"With the rapid development of computer vision, 3D data is increasing rapidly. How to retrieve similar model from a large number of models has become a hot research topic. However, in order to meet people's demand, the retrieval accuracy need to be further improved. In terms of multi-view 3D model retrieval, how to effectively learn the information between views is the key to improving performance. In this paper, we propose a novel 3D model retrieval algorithm based on attention and multi-view fusion. Specifically, we mainly constructed two modules. First, dynamic attentive graph learning module is used to learn the intrinsic relationship between view blocks; Then we propose the Attention-NetVlad algorithm, which combines the channel attention algorithm and the NetVlad algorithm. It learns the information between feature channels to enhance the feature expression ability firstly, then uses the NetVlad algorithm to fuse multiple view features into a global feature according to the clustering information. Finally the global feature is used as the only feature of the model to retrieve according to Euclidean distance. In comparison with other state-of-the-art methods by utilizing ModelNet10 and ModelNet40 the proposed method has demonstrated significant improvement for retrieval mAP. Our experiments also demonstrate the effectiveness of the modules in the algorithm.","PeriodicalId":145580,"journal":{"name":"Proceedings of the 5th International Conference on Computer Science and Software Engineering","volume":"17 17","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-10-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132644882","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}
In the task allocation of mobile crowdsensing (MCS), importance is often attached to the quality of sensing data while the data timeliness is often neglected, which may lead to the slow response of the MCS platform for urgent tasks (such as fire, geological disasters, etc.), thus missing the golden response time. Based on the definition of worker data timeliness and area coverage, a coverage-aware task allocation algorithm (CATA) is proposed in the paper. The CATA algorithm adopts fog nodes as the intermediate layer between MCS platform and participants and tries to both maximize the data timeliness and minimizing the incentive cost. For tasks with given location and crowdsensing range, workers with higher data timeliness and lower bidding are selected from participants according to their data timeliness and virtual credit. In addition, the location privacy of participants is protected by geo-indistinguishability. Results of simulation experiment validate the effectiveness of the proposed algorithm.
{"title":"A task allocation algorithm for Coverage-Aware Crowdsensing with Data timeliness","authors":"Chenxi Pan, Shuyu Li","doi":"10.1145/3569966.3570083","DOIUrl":"https://doi.org/10.1145/3569966.3570083","url":null,"abstract":"In the task allocation of mobile crowdsensing (MCS), importance is often attached to the quality of sensing data while the data timeliness is often neglected, which may lead to the slow response of the MCS platform for urgent tasks (such as fire, geological disasters, etc.), thus missing the golden response time. Based on the definition of worker data timeliness and area coverage, a coverage-aware task allocation algorithm (CATA) is proposed in the paper. The CATA algorithm adopts fog nodes as the intermediate layer between MCS platform and participants and tries to both maximize the data timeliness and minimizing the incentive cost. For tasks with given location and crowdsensing range, workers with higher data timeliness and lower bidding are selected from participants according to their data timeliness and virtual credit. In addition, the location privacy of participants is protected by geo-indistinguishability. Results of simulation experiment validate the effectiveness of the proposed algorithm.","PeriodicalId":145580,"journal":{"name":"Proceedings of the 5th International Conference on Computer Science and Software Engineering","volume":"65 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-10-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114965712","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}
Falls, considered a serious health-related concern for the elderly people, are associated with multiple diverse and dynamic needs for the elderly people themselves, their caregivers, their family members, and healthcare professionals. The modern-day Internet of Everything lifestyle is characterized by people using the internet for a multitude of reasons which also includes seeking and sharing information related to such needs. Such activity on the internet results in the generation of tremendous amounts of web behavior-based Big Data which can be studied and analyzed to investigate the trends in the underlining needs and the associated web search interests. The COVID-19 pandemic that the world is facing right now has impacted the elderly population to a significant extent. In fact, the elderly population is considered a demographic group that is most likely to get infected by this virus and develop serious symptoms, which could lead to hospitalizations and death. There hasn't been any study conducted in the field of aging research thus far that investigates how the COVID-19 pandemic may or may not have impacted the needs related to fall detection in the elderly. This work aims to address this research challenge. A dedicated methodology based on Google Trends is proposed in this paper that studies the web behavior-based Big Data related to fall detection from different countries both before and after the pandemic. The preliminary results presented from the analysis of the web behavior-based Big Data from 14 countries - USA, India, Germany, United Kingdom, Spain, Australia, Indonesia, Malaysia, Thailand, South Africa, Canada, Philippines, Sweden, and Ireland, which are amongst the countries worst hit by COVID-19, shows evidence that the pandemic had an impact towards increasing the web search interests related to fall detection in multiple countries.
{"title":"A Comprehensive Study to Analyze Trends in Web Search Interests Related to Fall Detection Before and After COVID-19","authors":"Nirmalya Thakur, Isabella Hall, Chia Y. Han","doi":"10.1145/3569966.3571193","DOIUrl":"https://doi.org/10.1145/3569966.3571193","url":null,"abstract":"Falls, considered a serious health-related concern for the elderly people, are associated with multiple diverse and dynamic needs for the elderly people themselves, their caregivers, their family members, and healthcare professionals. The modern-day Internet of Everything lifestyle is characterized by people using the internet for a multitude of reasons which also includes seeking and sharing information related to such needs. Such activity on the internet results in the generation of tremendous amounts of web behavior-based Big Data which can be studied and analyzed to investigate the trends in the underlining needs and the associated web search interests. The COVID-19 pandemic that the world is facing right now has impacted the elderly population to a significant extent. In fact, the elderly population is considered a demographic group that is most likely to get infected by this virus and develop serious symptoms, which could lead to hospitalizations and death. There hasn't been any study conducted in the field of aging research thus far that investigates how the COVID-19 pandemic may or may not have impacted the needs related to fall detection in the elderly. This work aims to address this research challenge. A dedicated methodology based on Google Trends is proposed in this paper that studies the web behavior-based Big Data related to fall detection from different countries both before and after the pandemic. The preliminary results presented from the analysis of the web behavior-based Big Data from 14 countries - USA, India, Germany, United Kingdom, Spain, Australia, Indonesia, Malaysia, Thailand, South Africa, Canada, Philippines, Sweden, and Ireland, which are amongst the countries worst hit by COVID-19, shows evidence that the pandemic had an impact towards increasing the web search interests related to fall detection in multiple countries.","PeriodicalId":145580,"journal":{"name":"Proceedings of the 5th International Conference on Computer Science and Software Engineering","volume":"190 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-10-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115010326","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}
Owing to the complex water environment, the acoustic point-cloud model formed by the detection method based on acoustic reflection mechanism is inevitably disturbed by the noise, which seriously affects the reconstruction effect of the underwater targets. Distinguishing between geometric features and noise is of paramount importance for the underwater point-cloud model filtering. Inspired by the classic image detail enhancement method of unsharp masking, we take the geometric coordinate information of the point as the research object and design a geometric feature-preserving adaptive unsharp masking filtering for the underwater point-cloud model. First, the proposed method directly performed a low-pass filtering using the neighborhood information to obtain the main structure of the input point-cloud model. Second, the detail layer was yielded by the difference between the input point-cloud model and the base layer. Third, the different scaling factors measuring the importance of the points with respect to the whole base layer were used to adaptively enhance the detail layer. Experimental results show that the proposed algorithm can effectively remove noise while maintaining the geometric characteristics of the model, which is obviously better than other comparison methods.
{"title":"Underwater Acoustic Point-cloud Filtering via Adaptive Unsharp Masking","authors":"Jisong Wang, Xuewu Zhang, Xiaolong Xu, Ke-Pu Song","doi":"10.1145/3569966.3570052","DOIUrl":"https://doi.org/10.1145/3569966.3570052","url":null,"abstract":"Owing to the complex water environment, the acoustic point-cloud model formed by the detection method based on acoustic reflection mechanism is inevitably disturbed by the noise, which seriously affects the reconstruction effect of the underwater targets. Distinguishing between geometric features and noise is of paramount importance for the underwater point-cloud model filtering. Inspired by the classic image detail enhancement method of unsharp masking, we take the geometric coordinate information of the point as the research object and design a geometric feature-preserving adaptive unsharp masking filtering for the underwater point-cloud model. First, the proposed method directly performed a low-pass filtering using the neighborhood information to obtain the main structure of the input point-cloud model. Second, the detail layer was yielded by the difference between the input point-cloud model and the base layer. Third, the different scaling factors measuring the importance of the points with respect to the whole base layer were used to adaptively enhance the detail layer. Experimental results show that the proposed algorithm can effectively remove noise while maintaining the geometric characteristics of the model, which is obviously better than other comparison methods.","PeriodicalId":145580,"journal":{"name":"Proceedings of the 5th International Conference on Computer Science and Software Engineering","volume":"9 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-10-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114932974","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}
Except for early surgical resection, melanoma lacks special treatment, while image segmentation can effectively assist doctors to enhance the efficiency of early diagnosis of melanoma. Due to the non-uniform size, shape and color of melanoma, it is difficult to segment the boundary of its lesion area. To solve the above problems, an improved DC-Unet network segmentation algorithm is proposed in this paper. A channel attention ECA-NET module was first introduced to make the model more focused on the lesion area of melanoma. Finally, the segmentation results are post-processed by Conditional Random Field (CRF) and Test Data Augmentation (TTA) to further refine the segmentation results. The experimental results showed that compared with the DC-Unet algorithm on the ISIC2017, ISIC2018 datasets, the segmentation accuracy was increased from 0.9513, 0.9444 to 0.9623, 0.9537 respectively.
{"title":"Skin melanoma segmentation algorithm using dual-channel efficient CNN network","authors":"Yadi Zhen, Jianbing Yi, Feng Cao, Jun Li, Jun Wu","doi":"10.1145/3569966.3570104","DOIUrl":"https://doi.org/10.1145/3569966.3570104","url":null,"abstract":"Except for early surgical resection, melanoma lacks special treatment, while image segmentation can effectively assist doctors to enhance the efficiency of early diagnosis of melanoma. Due to the non-uniform size, shape and color of melanoma, it is difficult to segment the boundary of its lesion area. To solve the above problems, an improved DC-Unet network segmentation algorithm is proposed in this paper. A channel attention ECA-NET module was first introduced to make the model more focused on the lesion area of melanoma. Finally, the segmentation results are post-processed by Conditional Random Field (CRF) and Test Data Augmentation (TTA) to further refine the segmentation results. The experimental results showed that compared with the DC-Unet algorithm on the ISIC2017, ISIC2018 datasets, the segmentation accuracy was increased from 0.9513, 0.9444 to 0.9623, 0.9537 respectively.","PeriodicalId":145580,"journal":{"name":"Proceedings of the 5th International Conference on Computer Science and Software Engineering","volume":"43 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-10-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122538164","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}
The perceptual hashing (pHash) algorithm generate a unique sequence of image. The similarity of images can be determined by comparing the distance between the hash sequences. A novel pHash methods is proposed in this paper.Firstly, the image after pre-processing is decomposed by NSCT into high-frequency and low-frequency parts, and the Zernike moments of high-frequency and LBP features of low-frequency are extracted. Secondly, extract the perceptual hashing features of the pre-processing image by using the LoG operator. Finally, the three feature sequences are concatenated to obtain the hash sequence of the image. Experimental results show that the proposed method outperforms other popular pHash algorithms in terms of uniqueness, differentiation, and robustness which means it can improve the effect of image retrieval.
{"title":"A Novel Image perceptual hashing algorithm based on frequency decomposition and LoG","authors":"Zihao Yang, Guosheng Hao, Xiaoyun Zhou, Wang Ruan","doi":"10.1145/3569966.3570057","DOIUrl":"https://doi.org/10.1145/3569966.3570057","url":null,"abstract":"The perceptual hashing (pHash) algorithm generate a unique sequence of image. The similarity of images can be determined by comparing the distance between the hash sequences. A novel pHash methods is proposed in this paper.Firstly, the image after pre-processing is decomposed by NSCT into high-frequency and low-frequency parts, and the Zernike moments of high-frequency and LBP features of low-frequency are extracted. Secondly, extract the perceptual hashing features of the pre-processing image by using the LoG operator. Finally, the three feature sequences are concatenated to obtain the hash sequence of the image. Experimental results show that the proposed method outperforms other popular pHash algorithms in terms of uniqueness, differentiation, and robustness which means it can improve the effect of image retrieval.","PeriodicalId":145580,"journal":{"name":"Proceedings of the 5th International Conference on Computer Science and Software Engineering","volume":"19 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-10-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125545589","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}
ABSTRACT: In the application of various fields of cryptography, the generation of the key should be random, because it directly determines the security of the system, so the detection of the random generation of the key is a crucial work in cryptography. Different detection methods detect different properties of sequences. The first interval test of non-overlapping templates introduced in this paper is mainly aimed at testing the frequency of the first interval of the specified template in the sequence. When the frequency of the first interval of the specified template conforms to the ideal distribution, the randomness hypothesis is accepted. On the contrary, the specific frequency the template, random hypothesis is rejected. This method focuses more on specifying the interval frequency of the occurrence of the template than some methods of checking the frequency of the occurrence of the template in the whole sequence. In addition, the sequence length examined by this detection method can be short or long. Compared with the random number detection method, the method introduced in this paper is more inclusive.
{"title":"A new method for detecting binary random sequences in cryptography: First interval test for non-overlapping templates","authors":"Xia Wu, Sheng Lin","doi":"10.1145/3569966.3569968","DOIUrl":"https://doi.org/10.1145/3569966.3569968","url":null,"abstract":"ABSTRACT: In the application of various fields of cryptography, the generation of the key should be random, because it directly determines the security of the system, so the detection of the random generation of the key is a crucial work in cryptography. Different detection methods detect different properties of sequences. The first interval test of non-overlapping templates introduced in this paper is mainly aimed at testing the frequency of the first interval of the specified template in the sequence. When the frequency of the first interval of the specified template conforms to the ideal distribution, the randomness hypothesis is accepted. On the contrary, the specific frequency the template, random hypothesis is rejected. This method focuses more on specifying the interval frequency of the occurrence of the template than some methods of checking the frequency of the occurrence of the template in the whole sequence. In addition, the sequence length examined by this detection method can be short or long. Compared with the random number detection method, the method introduced in this paper is more inclusive.","PeriodicalId":145580,"journal":{"name":"Proceedings of the 5th International Conference on Computer Science and Software Engineering","volume":"734 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-10-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116109801","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}
Power system short-term load forecasting plays an important role in the reliable, safe and economic operation of power system. Power system load forecasting data is an important basis for power grid planning, scheduling, marketing and other departments. In order to fully mine the effective information in the load data of power system and carry out accurate short-term load forecasting, this paper proposes a Long Short-Term Memory (LSTM) model based on wavelet denoising to build a short-term load forecasting model. Wavelet denoising method is used for data preprocessing, so as to ensure the accuracy of the prediction model, while LSTM is used to achieve high-quality short-term load forecasting of the power system. The method proposed in this paper has the advantages of strong training and learning ability, fast convergence speed, high prediction accuracy and strong adaptability.
{"title":"Research on Short-term Load Forecasting of Power System Based on Wavelet Denoising and Artificial Neural Network","authors":"Zihan Liu","doi":"10.1145/3569966.3569982","DOIUrl":"https://doi.org/10.1145/3569966.3569982","url":null,"abstract":"Power system short-term load forecasting plays an important role in the reliable, safe and economic operation of power system. Power system load forecasting data is an important basis for power grid planning, scheduling, marketing and other departments. In order to fully mine the effective information in the load data of power system and carry out accurate short-term load forecasting, this paper proposes a Long Short-Term Memory (LSTM) model based on wavelet denoising to build a short-term load forecasting model. Wavelet denoising method is used for data preprocessing, so as to ensure the accuracy of the prediction model, while LSTM is used to achieve high-quality short-term load forecasting of the power system. The method proposed in this paper has the advantages of strong training and learning ability, fast convergence speed, high prediction accuracy and strong adaptability.","PeriodicalId":145580,"journal":{"name":"Proceedings of the 5th International Conference on Computer Science and Software Engineering","volume":"81 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-10-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126343244","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}
Aimed at the lack of the spectral information preservation and the spatial detail injection in fusion of multispectral (MS) and panchromatic (PAN) images, the paper proposed a pansharpening algorithm based on convolutional sparse representation (CSR) and morphological filter (MF) by introducing a recently emerged signal decomposition model known as CSR. Firstly, the PAN and MS images are decomposed to obtain a base layer and a detail layer, respectively. Secondly, the fusion rule of the base layers which based on MF and high-pass modulation (HPM) scheme is proposed to retain more details. For the fusion of detail layers, maximum selection scheme based on activity maps and CSR model are adopted for fusion. Finally, the fusion results of the base layer and detail layer are reconstructed to obtain the final fusion image. The experimental results show that the proposed method is superior to the traditional methods and some current popular fusion methods from the visual effects and the objective indices.
{"title":"Fusion of multispectral and panchromatic images via convolutional sparse representation and morphological filter","authors":"Jiao Jiao, Depeng Chen, Shaobo Yu, Xin Guo","doi":"10.1145/3569966.3570103","DOIUrl":"https://doi.org/10.1145/3569966.3570103","url":null,"abstract":"Aimed at the lack of the spectral information preservation and the spatial detail injection in fusion of multispectral (MS) and panchromatic (PAN) images, the paper proposed a pansharpening algorithm based on convolutional sparse representation (CSR) and morphological filter (MF) by introducing a recently emerged signal decomposition model known as CSR. Firstly, the PAN and MS images are decomposed to obtain a base layer and a detail layer, respectively. Secondly, the fusion rule of the base layers which based on MF and high-pass modulation (HPM) scheme is proposed to retain more details. For the fusion of detail layers, maximum selection scheme based on activity maps and CSR model are adopted for fusion. Finally, the fusion results of the base layer and detail layer are reconstructed to obtain the final fusion image. The experimental results show that the proposed method is superior to the traditional methods and some current popular fusion methods from the visual effects and the objective indices.","PeriodicalId":145580,"journal":{"name":"Proceedings of the 5th International Conference on Computer Science and Software Engineering","volume":"7 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-10-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126098501","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：