With the rapid development of information technology, e-commerce has gradually entered thousands of households from high technology. E-commerce website (ECW) have become an important bridge for daily communication between enterprises and between enterprises and consumers. Image is one of the most important and commonly used information of human beings; Digital IPT is the process of converting image signal into digital format and processing it by computer. This paper studies and analyzes the application of graphics and IPT in the construction of ECW, discusses the graphics and IPT and the result design process of ECW, and tests the network name's views on the website construction proposed in this paper through online questionnaire experiment. The results show that 84% of Internet users using graphics and IPT in the construction of ECWs will make people more willing to browse websites, and 76% said that the visual effect of web pages is good, while websites without graphics and IPT have little effect on improving shopping desire, and only 61% of Internet users said they are willing to browse websites. It is verified that the application of graphics and IPT in the construction of ECW in this paper is conducive to improve the browsing volume of the website, and can also improve the visual aesthetic sense for Internet users, which is welcomed by the audience.
{"title":"E-commerce Website based on Graphics and Image Processing Technology (IPT)","authors":"Mihua Dang, Hongli Liu","doi":"10.1145/3561518.3561521","DOIUrl":"https://doi.org/10.1145/3561518.3561521","url":null,"abstract":"With the rapid development of information technology, e-commerce has gradually entered thousands of households from high technology. E-commerce website (ECW) have become an important bridge for daily communication between enterprises and between enterprises and consumers. Image is one of the most important and commonly used information of human beings; Digital IPT is the process of converting image signal into digital format and processing it by computer. This paper studies and analyzes the application of graphics and IPT in the construction of ECW, discusses the graphics and IPT and the result design process of ECW, and tests the network name's views on the website construction proposed in this paper through online questionnaire experiment. The results show that 84% of Internet users using graphics and IPT in the construction of ECWs will make people more willing to browse websites, and 76% said that the visual effect of web pages is good, while websites without graphics and IPT have little effect on improving shopping desire, and only 61% of Internet users said they are willing to browse websites. It is verified that the application of graphics and IPT in the construction of ECW in this paper is conducive to improve the browsing volume of the website, and can also improve the visual aesthetic sense for Internet users, which is welcomed by the audience.","PeriodicalId":196224,"journal":{"name":"Proceedings of the 6th International Conference on Graphics and Signal Processing","volume":"26 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116858251","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}
F. Bayram, Florian Pütz, Julian Weiß, R. Radtke, Alexander Jesser, N. Stache
This paper involves the development of an intelligent delineator for road traffic detecting potential conflict situations between motor vehicles and vulnerable road users at an early stage. By emitting warning signals, collisions between the road users concerned can then be prevented. The prototype used here includes, among other sensors, a high-resolution FMCW radar capable of detecting, and imaging objects. The goal of this work is to develop a Machine Learning (ML) model for object classification of vulnerable road users in radar frames. A 77 GHz chirp-sequence radar is used to record Range-Doppler maps from object classes of car, bicyclist, pedestrian and empty street at different locations. Objective of this is to cover different levels of background noise in the data caused by the different environments due to trees or bushes. For the data acquisition, simple traffic scenarios have been simulated at Heilbronn University. In selecting a suitable ML algorithm for the classifier, the main challenge was that modern machine learning methods are data-based models which require a lot of data and are generally lacking in explainability, such as neural networks. However, the great advantage is that the correlations in the data are learned automatically. With knowledge-based methods, on the other hand, the big advantage is that they are explainable and require much less data, but assume an extensive domain knowledge. Hybrid learning, also called Informed ML, represents a combination of the methods previously mentioned and their advantages. In this paper, one approach from each of these methods is selected as well as trained, and its results are compared to each other. The respective approaches investigated are a deep neural network (DNN), a Support Vector Machine (SVM), and a hybrid model of a SVM and a specific neural network for feature extraction called Autoencoder (AE). In this comparison the SVM performs with prediction accuracies around 80%. The hybrid model performs better achieving prediction accuracies around 90%. The best results of this comparator are achieved by the DNN, which has a prediction accuracy of around 98%.
{"title":"Classification of Vulnerable Road Users based on Range-Doppler Maps of 77 GHz MIMO Radar using Different Machine Learning Approaches","authors":"F. Bayram, Florian Pütz, Julian Weiß, R. Radtke, Alexander Jesser, N. Stache","doi":"10.1145/3561518.3561523","DOIUrl":"https://doi.org/10.1145/3561518.3561523","url":null,"abstract":"This paper involves the development of an intelligent delineator for road traffic detecting potential conflict situations between motor vehicles and vulnerable road users at an early stage. By emitting warning signals, collisions between the road users concerned can then be prevented. The prototype used here includes, among other sensors, a high-resolution FMCW radar capable of detecting, and imaging objects. The goal of this work is to develop a Machine Learning (ML) model for object classification of vulnerable road users in radar frames. A 77 GHz chirp-sequence radar is used to record Range-Doppler maps from object classes of car, bicyclist, pedestrian and empty street at different locations. Objective of this is to cover different levels of background noise in the data caused by the different environments due to trees or bushes. For the data acquisition, simple traffic scenarios have been simulated at Heilbronn University. In selecting a suitable ML algorithm for the classifier, the main challenge was that modern machine learning methods are data-based models which require a lot of data and are generally lacking in explainability, such as neural networks. However, the great advantage is that the correlations in the data are learned automatically. With knowledge-based methods, on the other hand, the big advantage is that they are explainable and require much less data, but assume an extensive domain knowledge. Hybrid learning, also called Informed ML, represents a combination of the methods previously mentioned and their advantages. In this paper, one approach from each of these methods is selected as well as trained, and its results are compared to each other. The respective approaches investigated are a deep neural network (DNN), a Support Vector Machine (SVM), and a hybrid model of a SVM and a specific neural network for feature extraction called Autoencoder (AE). In this comparison the SVM performs with prediction accuracies around 80%. The hybrid model performs better achieving prediction accuracies around 90%. The best results of this comparator are achieved by the DNN, which has a prediction accuracy of around 98%.","PeriodicalId":196224,"journal":{"name":"Proceedings of the 6th International Conference on Graphics and Signal Processing","volume":"103 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124576777","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 ecological protection red line includes areas with extremely important ecological functions such as biodiversity conservation and areas with extremely fragile ecology. The delineation of biodiversity protection and ecological protection red lines has a high degree of strategic fit, goal synergy and spatial consistency. Based on multi-source satellite data, this paper uses 3S technology as a means to construct a remote sensing biodiversity conservation function index, and conduct remote sensing monitoring and evaluation of the biodiversity conservation function of the ecological protection red line of Ruoergai wetland from 2000 to 2019. The protection of ecological protection red lines plays an important guiding role. The results show that: in terms of time, the area of the region with poor and poor biodiversity conservation function levels continues to shrink, and the area of the region above good continues to increase., the 20-year average biodiversity conservation function index evaluation grades accounted for 0.4% excellent, 18.2% good, 77.2% general, 3.9% poor, and 0.3% worst, mainly general and good, accounting for 95.4%, the grades of poor and worst are mainly distributed in the southeastern and northern parts of the red line area. In the past 20 years, the functional quality of biodiversity in the red line area of the Ruoergai Wetland has been maintained in a relatively good state, and the improvement in the northwest is more obvious than that in the southeast.
{"title":"Evaluation of Biodiversity Conservation Function of Ruoergai Wetland Ecological Protection Red Line Based on Satellite Data","authors":"Jie Yang, Yue Yuan, Bowei Wang","doi":"10.1145/3561518.3561530","DOIUrl":"https://doi.org/10.1145/3561518.3561530","url":null,"abstract":"The ecological protection red line includes areas with extremely important ecological functions such as biodiversity conservation and areas with extremely fragile ecology. The delineation of biodiversity protection and ecological protection red lines has a high degree of strategic fit, goal synergy and spatial consistency. Based on multi-source satellite data, this paper uses 3S technology as a means to construct a remote sensing biodiversity conservation function index, and conduct remote sensing monitoring and evaluation of the biodiversity conservation function of the ecological protection red line of Ruoergai wetland from 2000 to 2019. The protection of ecological protection red lines plays an important guiding role. The results show that: in terms of time, the area of the region with poor and poor biodiversity conservation function levels continues to shrink, and the area of the region above good continues to increase., the 20-year average biodiversity conservation function index evaluation grades accounted for 0.4% excellent, 18.2% good, 77.2% general, 3.9% poor, and 0.3% worst, mainly general and good, accounting for 95.4%, the grades of poor and worst are mainly distributed in the southeastern and northern parts of the red line area. In the past 20 years, the functional quality of biodiversity in the red line area of the Ruoergai Wetland has been maintained in a relatively good state, and the improvement in the northwest is more obvious than that in the southeast.","PeriodicalId":196224,"journal":{"name":"Proceedings of the 6th International Conference on Graphics and Signal Processing","volume":"26 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121905966","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}
Tourism 4.0 has become a trend increasing in the tourism sector in several countries. Many countries have prepared for the development of Tourism 4.0. Tourism 4.0 itself should benefit from technological developments from industry 4.0, such as IoT (Internet of Things), Big Data, Augmented Reality (AR), Virtual Reality (VR), Technology-based Business Models, Mobile Technology, Artificial Intelligent (AI)). One of the most critical technologies in the Tour and Travel industry is social media and the Internet. This research aims to analyze social media analytics data in Tour and Travel Industry on the Instagram platform to examine the developments of Tour and Travel companies after promotions have been performed on Instagram. The platform uses a social media analytics tool, analisa.io. It is an AI-powered social analytics software-as-a-service (SaaS) that provides platforms like Instagram and TikTok.
{"title":"Social Media Analysis in Tour and Travel Industry","authors":"Yohannes Kurniawan, I. Nugraha","doi":"10.1145/3561518.3561528","DOIUrl":"https://doi.org/10.1145/3561518.3561528","url":null,"abstract":"Tourism 4.0 has become a trend increasing in the tourism sector in several countries. Many countries have prepared for the development of Tourism 4.0. Tourism 4.0 itself should benefit from technological developments from industry 4.0, such as IoT (Internet of Things), Big Data, Augmented Reality (AR), Virtual Reality (VR), Technology-based Business Models, Mobile Technology, Artificial Intelligent (AI)). One of the most critical technologies in the Tour and Travel industry is social media and the Internet. This research aims to analyze social media analytics data in Tour and Travel Industry on the Instagram platform to examine the developments of Tour and Travel companies after promotions have been performed on Instagram. The platform uses a social media analytics tool, analisa.io. It is an AI-powered social analytics software-as-a-service (SaaS) that provides platforms like Instagram and TikTok.","PeriodicalId":196224,"journal":{"name":"Proceedings of the 6th International Conference on Graphics and Signal Processing","volume":"81 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130199777","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 continuous development of science and technology, people's requirements for the field of science and technology are getting higher and higher. The development of science and technology has brought more progress to the medical industry, and X-ray has made great contributions in all aspects. This paper firstly expounds the X-ray pathological examination instrument of the tomographic data, and then analyzes the image reconstruction algorithm, including the back-projection method and the filtered back-projection method. Finally, the physical basis of CT detection based on soft X-ray pathological detection equipment is studied, mainly the attenuation law of X-ray and the interaction between X-ray and matter. These findings indicate a major breakthrough in 3D reconstruction algorithms based on soft X-ray pathology scanners. This is of great help to the progress of medicine.
{"title":"3D Reconstruction Algorithm Based on Soft X-ray Pathology Examiner","authors":"Jun Yao, Yingyan Dou, Jingjuan Fu","doi":"10.1145/3561518.3561522","DOIUrl":"https://doi.org/10.1145/3561518.3561522","url":null,"abstract":"With the continuous development of science and technology, people's requirements for the field of science and technology are getting higher and higher. The development of science and technology has brought more progress to the medical industry, and X-ray has made great contributions in all aspects. This paper firstly expounds the X-ray pathological examination instrument of the tomographic data, and then analyzes the image reconstruction algorithm, including the back-projection method and the filtered back-projection method. Finally, the physical basis of CT detection based on soft X-ray pathological detection equipment is studied, mainly the attenuation law of X-ray and the interaction between X-ray and matter. These findings indicate a major breakthrough in 3D reconstruction algorithms based on soft X-ray pathology scanners. This is of great help to the progress of medicine.","PeriodicalId":196224,"journal":{"name":"Proceedings of the 6th International Conference on Graphics and Signal Processing","volume":"27 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127598235","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}
Word computation is a type of sentiment analysis that requires the identification not only of linguistic features, but also of the correlations of these features. There has been a great deal of research in this area. In order to understand linguistic representations and their applications in various domains of analysis, various factors such as demographics, emotions, and gender are taken into account in a transactional context. In this paper, we focus on those factors that can be extracted from existing data using natural language processing. We find that the most successful personality trait prediction models rely heavily on NLP techniques. To automate this process, researchers around the world have used a variety of machine learning and deep learning techniques. Different combinations of factors have led to different research results. We have conducted a comparative analysis of these experiments in the hope of determining the future course of action.
{"title":"Optimization of Language Models by Word Computing","authors":"Ka‐Hou Chan, S. Im, Yunfeng Zhang","doi":"10.1145/3561518.3561525","DOIUrl":"https://doi.org/10.1145/3561518.3561525","url":null,"abstract":"Word computation is a type of sentiment analysis that requires the identification not only of linguistic features, but also of the correlations of these features. There has been a great deal of research in this area. In order to understand linguistic representations and their applications in various domains of analysis, various factors such as demographics, emotions, and gender are taken into account in a transactional context. In this paper, we focus on those factors that can be extracted from existing data using natural language processing. We find that the most successful personality trait prediction models rely heavily on NLP techniques. To automate this process, researchers around the world have used a variety of machine learning and deep learning techniques. Different combinations of factors have led to different research results. We have conducted a comparative analysis of these experiments in the hope of determining the future course of action.","PeriodicalId":196224,"journal":{"name":"Proceedings of the 6th International Conference on Graphics and Signal Processing","volume":"13 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122363079","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}
Bi-LSTM-CRF (Bi-Directional Long Short-Term Memory Conditional Random Field) model have good performance in Chinese medical Electronic Medical Records (EMRS) Named Entity Recognition (NER), However, Bi-LSTM-CRF model cannot make full use of the parallelism of GPU (Graphics Processing Unit) in massive medical records, and the neglect of word order features and semantic information in IDCNN(Iterated Dilated Convolutional Neural Networks) model leads to poor NER effect. Therefore, this paper proposes a BERT-IDCNN-CRF model. In this model, the two-way transformer pre training model BERT is used to fine tune the model parameters in the manual annotated corpus conforming to the BIOES (Begin Inside Outside End Single) standard. The text is learned in an unsupervised manner, and the semantic information of words is represented by word vectors, which can well represent the context semantics in the sentences of EMRS; The state characteristics of character sequences are learned through BERT model, and the sequence state scores obtained are input to the CRF layer. The CRF layer makes constraint optimization on the sequence state transition, and IDCNN has better recognition effect on convolutional coding of local entities. Experimental test results: the average accuracy, recall and F1 value of the BERT-IDCNN-CRF model are 94.5%, 93.8% and 94.1% respectively, which are increased by 4.8%, 4.3% and 3.6% respectively compared with the baseline model Word2Vec-BiLSTM-CRF. The experiment proves that the BERT-IDCNN-CRF model can better identify medical entities in electronic medical records.
Bi-LSTM-CRF(双向长短期记忆条件随航场)模型在中国医疗电子病历(EMRS)命名实体识别(NER)中表现良好，但Bi-LSTM-CRF模型不能充分利用海量病历中GPU(图形处理单元)的并行性，且IDCNN(迭代扩张卷积神经网络)模型忽略了词序特征和语义信息，导致NER效果不佳。为此，本文提出了BERT-IDCNN-CRF模型。在该模型中，使用双向变压器预训练模型BERT对符合BIOES (Begin Inside Outside End Single)标准的手动标注语料库中的模型参数进行微调。以无监督的方式学习文本，用词向量表示词的语义信息，可以很好地表示EMRS句子中的上下文语义;通过BERT模型学习字符序列的状态特征，得到的序列状态分数输入到CRF层。CRF层对序列状态转换进行了约束优化，IDCNN对局部实体的卷积编码有较好的识别效果。实验测试结果:BERT-IDCNN-CRF模型的平均准确率、召回率和F1值分别为94.5%、93.8%和94.1%，比基线模型Word2Vec-BiLSTM-CRF分别提高了4.8%、4.3%和3.6%。实验证明BERT-IDCNN-CRF模型能较好地识别电子病历中的医疗实体。
{"title":"Research on Application of Named Entity Recognition of Electronic Medical Records Based on BERT-IDCNN-CRF Model","authors":"Xiaocheng Cai, Erhua Sun, Jiali Lei","doi":"10.1145/3561518.3561531","DOIUrl":"https://doi.org/10.1145/3561518.3561531","url":null,"abstract":"Bi-LSTM-CRF (Bi-Directional Long Short-Term Memory Conditional Random Field) model have good performance in Chinese medical Electronic Medical Records (EMRS) Named Entity Recognition (NER), However, Bi-LSTM-CRF model cannot make full use of the parallelism of GPU (Graphics Processing Unit) in massive medical records, and the neglect of word order features and semantic information in IDCNN(Iterated Dilated Convolutional Neural Networks) model leads to poor NER effect. Therefore, this paper proposes a BERT-IDCNN-CRF model. In this model, the two-way transformer pre training model BERT is used to fine tune the model parameters in the manual annotated corpus conforming to the BIOES (Begin Inside Outside End Single) standard. The text is learned in an unsupervised manner, and the semantic information of words is represented by word vectors, which can well represent the context semantics in the sentences of EMRS; The state characteristics of character sequences are learned through BERT model, and the sequence state scores obtained are input to the CRF layer. The CRF layer makes constraint optimization on the sequence state transition, and IDCNN has better recognition effect on convolutional coding of local entities. Experimental test results: the average accuracy, recall and F1 value of the BERT-IDCNN-CRF model are 94.5%, 93.8% and 94.1% respectively, which are increased by 4.8%, 4.3% and 3.6% respectively compared with the baseline model Word2Vec-BiLSTM-CRF. The experiment proves that the BERT-IDCNN-CRF model can better identify medical entities in electronic medical records.","PeriodicalId":196224,"journal":{"name":"Proceedings of the 6th International Conference on Graphics and Signal Processing","volume":"4 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133048307","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}
Lanqing Ou, Dawa Pengcuo, Tudeng Jiangcuo, Min Zhao
There are 12 nasal consonants in Tibetan Yushu dialect, involving three phonation types and four places of articulation. This paper explores Yushu Tibetan nasals with Nasalance Visualization System and finds that for the nasals with the same place of articulation, the nasalance score always presents a common rule: voiceless voice >modal voice> anterior-glottal stop voice. We also find that the nasalance score of the nasal initials with different places of articulation is affected differently by the subsequent vowel: for the nasal initials in the front, that is, for the bilabial nasals and dental nasals, the higher the position of the tongue of the following vowel, the higher the nasalance score of the nasal; for the nasal initials at the back, namely for the alveolar- palatal nasals and velar nasals, the more forward the position of the following vowel, the higher the nasalance score of the nasal. We also find that the two nasal finals are totally different: the nasalance score of final /ŋ/ is always high, while the nasalance score of the final /n/ is unstable, varying from 65 to 95, which indicates that the final /n/ is weakening.
{"title":"Analyses of Tibetan Yushu Dialect Nasals Based on Nasalance Visualization System","authors":"Lanqing Ou, Dawa Pengcuo, Tudeng Jiangcuo, Min Zhao","doi":"10.1145/3561518.3561526","DOIUrl":"https://doi.org/10.1145/3561518.3561526","url":null,"abstract":"There are 12 nasal consonants in Tibetan Yushu dialect, involving three phonation types and four places of articulation. This paper explores Yushu Tibetan nasals with Nasalance Visualization System and finds that for the nasals with the same place of articulation, the nasalance score always presents a common rule: voiceless voice >modal voice> anterior-glottal stop voice. We also find that the nasalance score of the nasal initials with different places of articulation is affected differently by the subsequent vowel: for the nasal initials in the front, that is, for the bilabial nasals and dental nasals, the higher the position of the tongue of the following vowel, the higher the nasalance score of the nasal; for the nasal initials at the back, namely for the alveolar- palatal nasals and velar nasals, the more forward the position of the following vowel, the higher the nasalance score of the nasal. We also find that the two nasal finals are totally different: the nasalance score of final /ŋ/ is always high, while the nasalance score of the final /n/ is unstable, varying from 65 to 95, which indicates that the final /n/ is weakening.","PeriodicalId":196224,"journal":{"name":"Proceedings of the 6th International Conference on Graphics and Signal Processing","volume":"24 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122573767","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}
Face expression plays a critical role during the daily life, and people cannot live without face emotion. With the development of technology, many methods of facial expression recognition have been proposed. However, from traditional methods to deep learning methods, few of them pay attention to the hybrid data augmentation, which can help improve the robustness of models. Therefore, a method of hybrid data augmentation is highlighted in this paper. The hybrid data augmentation is a method of combining several effective data augmentation. In the experiments, the technique is applied on four basic networks and the results are compared to the baseline models. After applying this technique, the results show that four benchmark models have higher performance than those previously. This approach is simple and robust in terms of data augmentation, which makes it applicated in the real world in the future. Besides the results show versatility of the technique as all of our experiments get better results.
{"title":"Face Emotion Recognization Using Dataset Augmentation Based on Neural Network","authors":"M. Rao, Ruying Bao, Liangshun Dong","doi":"10.1145/3561518.3561519","DOIUrl":"https://doi.org/10.1145/3561518.3561519","url":null,"abstract":"Face expression plays a critical role during the daily life, and people cannot live without face emotion. With the development of technology, many methods of facial expression recognition have been proposed. However, from traditional methods to deep learning methods, few of them pay attention to the hybrid data augmentation, which can help improve the robustness of models. Therefore, a method of hybrid data augmentation is highlighted in this paper. The hybrid data augmentation is a method of combining several effective data augmentation. In the experiments, the technique is applied on four basic networks and the results are compared to the baseline models. After applying this technique, the results show that four benchmark models have higher performance than those previously. This approach is simple and robust in terms of data augmentation, which makes it applicated in the real world in the future. Besides the results show versatility of the technique as all of our experiments get better results.","PeriodicalId":196224,"journal":{"name":"Proceedings of the 6th International Conference on Graphics and Signal Processing","volume":"6 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122233835","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}
Compared with conventional communication systems, spread spectrum communication system has the advantages of anti-interference, multiple access multiplexing, strong confidentiality and so on. Therefore, spread spectrum communication has been widely used in civil and military fields. Acquisition is one of the core technologies of spread spectrum communication system. How to capture spread spectrum signal in the case of low signal-to-noise ratio and large Doppler frequency offset becomes the most critical link for the establishment of spread spectrum system. This scheme adopts a two-dimensional acquisition method based on FFT, which can combine time-domain search and frequency-domain search, and provide a two-dimensional spread spectrum signal acquisition method with high-precision through matched filter.
{"title":"Research on Spread Spectrum Signal Acquisition","authors":"Yuchen Li, Feng Tian, Mengjia Ge, Chuangxin Zhao, Xiang Cao, Qianya Lu, Hanghua Gan","doi":"10.1145/3561518.3561527","DOIUrl":"https://doi.org/10.1145/3561518.3561527","url":null,"abstract":"Compared with conventional communication systems, spread spectrum communication system has the advantages of anti-interference, multiple access multiplexing, strong confidentiality and so on. Therefore, spread spectrum communication has been widely used in civil and military fields. Acquisition is one of the core technologies of spread spectrum communication system. How to capture spread spectrum signal in the case of low signal-to-noise ratio and large Doppler frequency offset becomes the most critical link for the establishment of spread spectrum system. This scheme adopts a two-dimensional acquisition method based on FFT, which can combine time-domain search and frequency-domain search, and provide a two-dimensional spread spectrum signal acquisition method with high-precision through matched filter.","PeriodicalId":196224,"journal":{"name":"Proceedings of the 6th International Conference on Graphics and Signal Processing","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129668754","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：