Pub Date : 2021-08-01DOI: 10.1109/ISCEIC53685.2021.00064
Meng’An Shi, Huimin Cai, Yang Gao
This paper briefly describes the similarities and differences of the mainstream models of deep learning target detection box, analyzes the characteristics and advantages of Mask RCNN, a universal target detection box, and focuses on the application of Mask RCNN in human posture detection in multi- person human posture task. Through the analysis, it is considered that the advantage of Mask RCNN in multi-person human posture detection task is the accuracy, while the bottleneck is the detection speed. To solve this problem, an optimization of Mask RCNN model based on MobileNet was proposed to accelerate the inference calculation speed of Mask RCNN. At the same time, in order to further improve the detection accuracy of Mask RCNN, a method of using pixel segmentation results to assist the detection of human body key points is proposed. Experimental results show that compared with the original algorithm, it improves the reasoning speed and reduces the false detection rate caused by the environment.
{"title":"Optimization of Human Pose Detection Based on Mask RCNN","authors":"Meng’An Shi, Huimin Cai, Yang Gao","doi":"10.1109/ISCEIC53685.2021.00064","DOIUrl":"https://doi.org/10.1109/ISCEIC53685.2021.00064","url":null,"abstract":"This paper briefly describes the similarities and differences of the mainstream models of deep learning target detection box, analyzes the characteristics and advantages of Mask RCNN, a universal target detection box, and focuses on the application of Mask RCNN in human posture detection in multi- person human posture task. Through the analysis, it is considered that the advantage of Mask RCNN in multi-person human posture detection task is the accuracy, while the bottleneck is the detection speed. To solve this problem, an optimization of Mask RCNN model based on MobileNet was proposed to accelerate the inference calculation speed of Mask RCNN. At the same time, in order to further improve the detection accuracy of Mask RCNN, a method of using pixel segmentation results to assist the detection of human body key points is proposed. Experimental results show that compared with the original algorithm, it improves the reasoning speed and reduces the false detection rate caused by the environment.","PeriodicalId":342968,"journal":{"name":"2021 2nd International Symposium on Computer Engineering and Intelligent Communications (ISCEIC)","volume":"35 1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131695032","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 development of intelligent phones and speech recognition technology, there is a great demand for generating meeting minutes automatically. In this paper, we design and implement Yanji, an automated system for generating meeting minutes based on speech and speaker recognition. The Yanji system realizes the following functions: recording the audio of the conference, uploading audios to IBM cloud in real time, transcribing audios to texts and identifying various speakers with IBM Speech to Text API, and finally generating complete meeting minutes. Yanji greatly reduces the recording storage space and the labor cost of listening and writing, and improves the meeting efficiency.
随着智能手机和语音识别技术的发展，会议纪要的自动生成需求越来越大。本文设计并实现了基于语音和说话人识别的会议纪要自动生成系统“延吉”。延吉系统实现了以下功能:录制会议音频，将音频实时上传到IBM云，将音频转换为文本，并通过IBM Speech to Text API识别不同的演讲者，最终生成完整的会议纪要。延吉大大减少了录音存储空间和听写的人工成本，提高了会议效率。
{"title":"Yanji: An Automated Mobile Meeting Minutes System","authors":"Xuning Chen, Fengwei Sheng, Rongxuan He, Shiwei Chen, Hongmeng Ma, Yanfeng Wu, Jing Xu","doi":"10.1109/ISCEIC53685.2021.00095","DOIUrl":"https://doi.org/10.1109/ISCEIC53685.2021.00095","url":null,"abstract":"With the development of intelligent phones and speech recognition technology, there is a great demand for generating meeting minutes automatically. In this paper, we design and implement Yanji, an automated system for generating meeting minutes based on speech and speaker recognition. The Yanji system realizes the following functions: recording the audio of the conference, uploading audios to IBM cloud in real time, transcribing audios to texts and identifying various speakers with IBM Speech to Text API, and finally generating complete meeting minutes. Yanji greatly reduces the recording storage space and the labor cost of listening and writing, and improves the meeting efficiency.","PeriodicalId":342968,"journal":{"name":"2021 2nd International Symposium on Computer Engineering and Intelligent Communications (ISCEIC)","volume":"48 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128056234","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 : 2021-08-01DOI: 10.1109/ISCEIC53685.2021.00074
Wenxiang Zhang, Yingchun Zhong, Zongyang Wang, Zhiyong Luo, Bo Wang
It is one of the key steps of unmanned aerial vehicles (UAVs) autonomously cruising to touch down the distributed airport automatically and accurately. The ground-effect wind field is one of the main factors to influence the touching down precision of UAV. In order to explore how much the ground-effect wind field influence the touching down precision, this paper chooses a quadrotor UAV with 400mm diameter as the research object, establishes the model of UAV and the ground-effect wind field, analyzes the touching down results while the distributed airport is at the opening and flat scene, the scene with single-sided vertical obstacle and the scene with bi-symmetrical vertical obstacles respectively. The experimental results show that: (1) there is little affection on the accuracy of touching down when the distributed airport is at the opening and flat scene. (2) The ground-effect wind field has great affection on the accuracy of touching down when the distributed airport is at the scene with single-sided vertical obstacle. If the distance between the UAV and the single-sided vertical obstacle is less than 1 meter, the UAV is very easy to overturn. (3) When the distributed airport is at the scene with bi-symmetrical vertical obstacles, the accuracy of touching down is decided by the height of the vertical obstacle. The investigation of this paper is very significant to select the location of distributed airport.
{"title":"Influence on distributed airport site selection from ground-effect wind field of UAV","authors":"Wenxiang Zhang, Yingchun Zhong, Zongyang Wang, Zhiyong Luo, Bo Wang","doi":"10.1109/ISCEIC53685.2021.00074","DOIUrl":"https://doi.org/10.1109/ISCEIC53685.2021.00074","url":null,"abstract":"It is one of the key steps of unmanned aerial vehicles (UAVs) autonomously cruising to touch down the distributed airport automatically and accurately. The ground-effect wind field is one of the main factors to influence the touching down precision of UAV. In order to explore how much the ground-effect wind field influence the touching down precision, this paper chooses a quadrotor UAV with 400mm diameter as the research object, establishes the model of UAV and the ground-effect wind field, analyzes the touching down results while the distributed airport is at the opening and flat scene, the scene with single-sided vertical obstacle and the scene with bi-symmetrical vertical obstacles respectively. The experimental results show that: (1) there is little affection on the accuracy of touching down when the distributed airport is at the opening and flat scene. (2) The ground-effect wind field has great affection on the accuracy of touching down when the distributed airport is at the scene with single-sided vertical obstacle. If the distance between the UAV and the single-sided vertical obstacle is less than 1 meter, the UAV is very easy to overturn. (3) When the distributed airport is at the scene with bi-symmetrical vertical obstacles, the accuracy of touching down is decided by the height of the vertical obstacle. The investigation of this paper is very significant to select the location of distributed airport.","PeriodicalId":342968,"journal":{"name":"2021 2nd International Symposium on Computer Engineering and Intelligent Communications (ISCEIC)","volume":"15 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123025953","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 : 2021-08-01DOI: 10.1109/ISCEIC53685.2021.00088
Xin Zhen, Jinhua Li
In order to improve image clarity and ensure image processing effects, a global optimization image completion method based on generative confrontation network is proposed. The defect area of the image is collected and detected, and the feature changes of the globally optimized image are analyzed, thereby effectively de-noising the image information and effectively improving the image quality. Finally, experiments show that the global optimization image completion processing method based on the generative confrontation network can better improve the image definition and has high practicability.
{"title":"Global optimization image completion processing based on generative countermeasure network","authors":"Xin Zhen, Jinhua Li","doi":"10.1109/ISCEIC53685.2021.00088","DOIUrl":"https://doi.org/10.1109/ISCEIC53685.2021.00088","url":null,"abstract":"In order to improve image clarity and ensure image processing effects, a global optimization image completion method based on generative confrontation network is proposed. The defect area of the image is collected and detected, and the feature changes of the globally optimized image are analyzed, thereby effectively de-noising the image information and effectively improving the image quality. Finally, experiments show that the global optimization image completion processing method based on the generative confrontation network can better improve the image definition and has high practicability.","PeriodicalId":342968,"journal":{"name":"2021 2nd International Symposium on Computer Engineering and Intelligent Communications (ISCEIC)","volume":"121 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115209754","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 : 2021-08-01DOI: 10.1109/ISCEIC53685.2021.00082
Haiyang Zheng, Yingchun Zhong, Wenxiang Zhang, Zhiyong Luo, Bo Wang
It is one of the significant tasks of power inspection by multi rotor Unmanned Aerial Vehicle (UAV) to recognize engineering vehicles in aerial images. If there are engineering vehicles working near or below the high-voltage power line, the UAV would give out the important early warning message to avoid the situation that the bucket or boom of the engineering vehicle enters the safe distance from the high-voltage power line, and reduce accidents such as short circuit breakdown. Aiming at the problem of recognition of engineering vehicles in aerial images of UAV inspection, this paper proposed an improved capsule network method. First, the structure of original capsule network is replaced with a multi-layer densely connected capsule network. Next, the dynamic routing algorithm of the capsule network is improved. As the results of experiments have shown, (1) the improved capsule network method gets a mAP of 93.74% for the recognition of engineering vehicles, and its parameter scale is smaller than other methods. (2) The number of network layers influences the recognition precision greatly. Their relationship is non-monotonic and nonlinear. In addition, whether or not to improve the dynamic routing algorithm does not affect the trends of recognition mAP. The overall performance of the improved capsule network method is obviously better than YOLOv5 and other artificial feature extraction methods.
{"title":"Recognition of Engineering Vehicles in Aerial Images of Multi Rotor UAV","authors":"Haiyang Zheng, Yingchun Zhong, Wenxiang Zhang, Zhiyong Luo, Bo Wang","doi":"10.1109/ISCEIC53685.2021.00082","DOIUrl":"https://doi.org/10.1109/ISCEIC53685.2021.00082","url":null,"abstract":"It is one of the significant tasks of power inspection by multi rotor Unmanned Aerial Vehicle (UAV) to recognize engineering vehicles in aerial images. If there are engineering vehicles working near or below the high-voltage power line, the UAV would give out the important early warning message to avoid the situation that the bucket or boom of the engineering vehicle enters the safe distance from the high-voltage power line, and reduce accidents such as short circuit breakdown. Aiming at the problem of recognition of engineering vehicles in aerial images of UAV inspection, this paper proposed an improved capsule network method. First, the structure of original capsule network is replaced with a multi-layer densely connected capsule network. Next, the dynamic routing algorithm of the capsule network is improved. As the results of experiments have shown, (1) the improved capsule network method gets a mAP of 93.74% for the recognition of engineering vehicles, and its parameter scale is smaller than other methods. (2) The number of network layers influences the recognition precision greatly. Their relationship is non-monotonic and nonlinear. In addition, whether or not to improve the dynamic routing algorithm does not affect the trends of recognition mAP. The overall performance of the improved capsule network method is obviously better than YOLOv5 and other artificial feature extraction methods.","PeriodicalId":342968,"journal":{"name":"2021 2nd International Symposium on Computer Engineering and Intelligent Communications (ISCEIC)","volume":"47 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121138596","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 : 2021-08-01DOI: 10.1109/ISCEIC53685.2021.00080
Xiaoyu Cong, Pandong Zhang, Yubing Han
Jamming identification is the premise of radar anti-jamming in the complex electromagnetic environment. The signals from monostatic radar are taken as the object of training and identification, which has the disadvantages of less information, single observation angle and easy to be attacked. In order to improve the identification accuracy, a jamming identification method based on deep learning for networking radars is proposed in this paper. The range-Doppler signals from multiple radars in the network are stitched into a data set for jamming identification, which contains more information than that from monostatic radar. The models of radar jammings are established, and a Convolutional Neural Network is designed to identify jammings, target signal and noise. The simulation results show that the accuracy of the proposed jamming identification method is 99.2%.
{"title":"A jamming identification method based on deep learning for networking radars","authors":"Xiaoyu Cong, Pandong Zhang, Yubing Han","doi":"10.1109/ISCEIC53685.2021.00080","DOIUrl":"https://doi.org/10.1109/ISCEIC53685.2021.00080","url":null,"abstract":"Jamming identification is the premise of radar anti-jamming in the complex electromagnetic environment. The signals from monostatic radar are taken as the object of training and identification, which has the disadvantages of less information, single observation angle and easy to be attacked. In order to improve the identification accuracy, a jamming identification method based on deep learning for networking radars is proposed in this paper. The range-Doppler signals from multiple radars in the network are stitched into a data set for jamming identification, which contains more information than that from monostatic radar. The models of radar jammings are established, and a Convolutional Neural Network is designed to identify jammings, target signal and noise. The simulation results show that the accuracy of the proposed jamming identification method is 99.2%.","PeriodicalId":342968,"journal":{"name":"2021 2nd International Symposium on Computer Engineering and Intelligent Communications (ISCEIC)","volume":"9 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126496689","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 : 2021-08-01DOI: 10.1109/ISCEIC53685.2021.00086
Qin Zhu, Jun Tang, Jinwang Yi, Qiuyue Yin
Multiple-Input Multiple-Output (MIMO) radar can efficiently improve radar performance by transmitting specific orthogonal waveforms. A novel multi-pulse waveforms design is proposed for MIMO radar in this paper. The polyphase complementary sequences are used as spatial codes and then combined with circulating linear frequency modulated (LFM) signals to enhance range resolution in this method. Both analyses and results show that the designed waveform has better performance than the conventional signals in suppressing sidelobes and range resolution.
{"title":"Design of multi-pulse waveforms using polyphase complementary phase-coding in MIMO radar","authors":"Qin Zhu, Jun Tang, Jinwang Yi, Qiuyue Yin","doi":"10.1109/ISCEIC53685.2021.00086","DOIUrl":"https://doi.org/10.1109/ISCEIC53685.2021.00086","url":null,"abstract":"Multiple-Input Multiple-Output (MIMO) radar can efficiently improve radar performance by transmitting specific orthogonal waveforms. A novel multi-pulse waveforms design is proposed for MIMO radar in this paper. The polyphase complementary sequences are used as spatial codes and then combined with circulating linear frequency modulated (LFM) signals to enhance range resolution in this method. Both analyses and results show that the designed waveform has better performance than the conventional signals in suppressing sidelobes and range resolution.","PeriodicalId":342968,"journal":{"name":"2021 2nd International Symposium on Computer Engineering and Intelligent Communications (ISCEIC)","volume":"30 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132511589","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 : 2021-08-01DOI: 10.1109/ISCEIC53685.2021.00009
Haiyan Lv, Zhihong Zhang
With the explosive growth of mobile devices, the computing power and resource storage of the mobile terminal have higher demand, and the workflow task scheduling in the moving edge computing environment can solve complex data dependencies between tasks in workflows. The proposal of moving edge computing techniques is to solve computing resources caused by massive mobile device access, which can meet the low delay and high computing power of the mobile device. This paper proposes a workflow task scheduling optimization algorithm (WTS-OSM) in moving edge computing environment. First, the workflow tasks are generated into directed acyclic graph (DAG), and then the tasks in DAG are layered. The tasks in the same layer do not have dependencies, but the tasks in two adjacent layers do. Finally, an optimized genetic algorithm (GA) is used to determine whether the layer task computes unloading. Experimental results show that the proposed algorithm is superior to the traditional algorithms in terms of the task execution time.
{"title":"Workflow task scheduling optimization strategy in moving edge computing environment","authors":"Haiyan Lv, Zhihong Zhang","doi":"10.1109/ISCEIC53685.2021.00009","DOIUrl":"https://doi.org/10.1109/ISCEIC53685.2021.00009","url":null,"abstract":"With the explosive growth of mobile devices, the computing power and resource storage of the mobile terminal have higher demand, and the workflow task scheduling in the moving edge computing environment can solve complex data dependencies between tasks in workflows. The proposal of moving edge computing techniques is to solve computing resources caused by massive mobile device access, which can meet the low delay and high computing power of the mobile device. This paper proposes a workflow task scheduling optimization algorithm (WTS-OSM) in moving edge computing environment. First, the workflow tasks are generated into directed acyclic graph (DAG), and then the tasks in DAG are layered. The tasks in the same layer do not have dependencies, but the tasks in two adjacent layers do. Finally, an optimized genetic algorithm (GA) is used to determine whether the layer task computes unloading. Experimental results show that the proposed algorithm is superior to the traditional algorithms in terms of the task execution time.","PeriodicalId":342968,"journal":{"name":"2021 2nd International Symposium on Computer Engineering and Intelligent Communications (ISCEIC)","volume":"26 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130196699","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 : 2021-08-01DOI: 10.1109/ISCEIC53685.2021.00041
Honghao Yu, Hong Jia
In order to improve the storage and management ability of relational database for massive meteorological grid data, an index method based on spatial subdivision grid is proposed, and a distributed storage model of meteorological grid data based on HBase is designed and implemented. In this paper, the spatial grid is established on the NetCDF format meteorological grid data, and the spatial index coding based on spatial octree is designed; The distributed storage scheme based on HBase is adopted and the structure of rowkey and table is optimized . The experimental results show that HBase database is superior to relational database in storage and query for massive meteorological grid data.
{"title":"Distributed Storage of Meteorological Grid Data based on HBase","authors":"Honghao Yu, Hong Jia","doi":"10.1109/ISCEIC53685.2021.00041","DOIUrl":"https://doi.org/10.1109/ISCEIC53685.2021.00041","url":null,"abstract":"In order to improve the storage and management ability of relational database for massive meteorological grid data, an index method based on spatial subdivision grid is proposed, and a distributed storage model of meteorological grid data based on HBase is designed and implemented. In this paper, the spatial grid is established on the NetCDF format meteorological grid data, and the spatial index coding based on spatial octree is designed; The distributed storage scheme based on HBase is adopted and the structure of rowkey and table is optimized . The experimental results show that HBase database is superior to relational database in storage and query for massive meteorological grid data.","PeriodicalId":342968,"journal":{"name":"2021 2nd International Symposium on Computer Engineering and Intelligent Communications (ISCEIC)","volume":"67 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126652022","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 : 2021-08-01DOI: 10.1109/ISCEIC53685.2021.00052
Shida Zhang, Jingyu Zhang, Zehua Wang, Quanhu Li
Aiming at the problem that the common material grinding factors in the industry are complex and it is difficult to accurately predict the output particle size, this paper introduces the sparrow search algorithm, and proposes two improved strategies for the sparrow search algorithm. For the original sparrow search algorithm, the global search ability is insufficient. And the problem that is easy to fall into the local optimum, the introduction of Tent chaotic map to initialize the population, enhance the global search ability. Meanwhile, introduce the Cauchy mutation strategy to solve the local optimum problem, effectively improve the algorithm search ability, and combine the BP neural network to grind the output particle size of the material make predictions. The simulation results show that the improved sparrow search algorithm optimizes the weights and biases of the BP neural network and improves the training accuracy of the BP neural network. The experimental results show that the proposed TCSSA-BP model has obvious effects on the regression prediction of the output particle size of the material grinding.
{"title":"Regression prediction of material grinding particle size based on improved sparrow search algorithm to optimize BP neural network","authors":"Shida Zhang, Jingyu Zhang, Zehua Wang, Quanhu Li","doi":"10.1109/ISCEIC53685.2021.00052","DOIUrl":"https://doi.org/10.1109/ISCEIC53685.2021.00052","url":null,"abstract":"Aiming at the problem that the common material grinding factors in the industry are complex and it is difficult to accurately predict the output particle size, this paper introduces the sparrow search algorithm, and proposes two improved strategies for the sparrow search algorithm. For the original sparrow search algorithm, the global search ability is insufficient. And the problem that is easy to fall into the local optimum, the introduction of Tent chaotic map to initialize the population, enhance the global search ability. Meanwhile, introduce the Cauchy mutation strategy to solve the local optimum problem, effectively improve the algorithm search ability, and combine the BP neural network to grind the output particle size of the material make predictions. The simulation results show that the improved sparrow search algorithm optimizes the weights and biases of the BP neural network and improves the training accuracy of the BP neural network. The experimental results show that the proposed TCSSA-BP model has obvious effects on the regression prediction of the output particle size of the material grinding.","PeriodicalId":342968,"journal":{"name":"2021 2nd International Symposium on Computer Engineering and Intelligent Communications (ISCEIC)","volume":"47 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133583947","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：