2022 Asia Conference on Algorithms, Computing and Machine Learning (CACML)最新文献_第7页

Automatic Segmentation Method of Multiple Objects in Chip Package Outline Drawings 芯片封装轮廓图中多目标的自动分割方法

2022 Asia Conference on Algorithms, Computing and Machine Learning (CACML) Pub Date : 2022-03-01 DOI: 10.1109/CACML55074.2022.00041

Lingfei Tang, Xiaoyu Liang, Lv Wu, N. Xu

{"title":"Automatic Segmentation Method of Multiple Objects in Chip Package Outline Drawings","authors":"Lingfei Tang, Xiaoyu Liang, Lv Wu, N. Xu","doi":"10.1109/CACML55074.2022.00041","DOIUrl":"https://doi.org/10.1109/CACML55074.2022.00041","url":null,"abstract":"Chip package design is an important part of chip design automation, and chip package outline drawings are the basis for chip package design, which plays an important role. The automatic segmentation of the chip package outline drawing can improve the work efficiency of engineers. This paper proposes an effective technique for automatic segmentation of the chip package outline drawing, which realizes the segmentation and extraction of multiple objects in the chip package outline drawing. Mainly through image preprocessing, morphological processing, contour extraction, multi-objects segmentation and extraction four parts to achieve. The adaptive dilation method is proposed in the morphological processing, and the approximate circumscribed polygon method is proposed in the contour extraction, in the multi-objects segmentation part, a method of extracting objects according to irregular contour segmentation is proposed. The experimental results compared with other classical methods show that the method proposed in this paper works best. In the chip package outline drawing samples including 346 objects to be segmented, the method achieves a segmentation accuracy of 86.99% demonstrating the effectiveness of the proposed technique.","PeriodicalId":137505,"journal":{"name":"2022 Asia Conference on Algorithms, Computing and Machine Learning (CACML)","volume":"54 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126739990","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}

引用次数: 0

Propagation research of SARS-CoV-2 based on evolutionary tree and spectral clustering 基于进化树和谱聚类的SARS-CoV-2传播研究

2022 Asia Conference on Algorithms, Computing and Machine Learning (CACML) Pub Date : 2022-03-01 DOI: 10.1109/CACML55074.2022.00106

Jingyi Yan, Yong Cao, Naien Wu, Fei Xiong, Yongke Sun, Cong Zhong

{"title":"Propagation research of SARS-CoV-2 based on evolutionary tree and spectral clustering","authors":"Jingyi Yan, Yong Cao, Naien Wu, Fei Xiong, Yongke Sun, Cong Zhong","doi":"10.1109/CACML55074.2022.00106","DOIUrl":"https://doi.org/10.1109/CACML55074.2022.00106","url":null,"abstract":"RNA viruses have the characteristics of a high mutation rate. New Coronavirus (SARS-CoV-2), as a RNA virus, has been mutated to some extent since the outbreak of New Coronavirus pneumonia (COVID-19). It is of great significance to study the evolution and variation of novel coronavirus genes to analyze the source of virus infection and understand the evolution of viruses. This research is based on the Novel Coronavirus 2019 database at the National Genomics Data Center. We combined macro and micro. We used the phylogenetic tree to analyze the gene fragments of the virus, constructed an evolutionary tree with a depth of 301, searched the root node of the tree to find the source of the virus in the data set and used spectral clustering to analyze the degree of novel Coronavirus variation in each country and the clustering results were visualized to make them easier to observe. The experimental results show that the strain sample at the top of the evolutionary tree originated in New Zealand based on the existing data. In the evolutionary tree, the evolutionary process of the virus can be divided into three branches. After clustering the virus source data and constructing the visual map of the variation degree of SARS-COV-2, we found that the viruses in South Africa, New Zealand and other countries had a higher degree of variation, and the viruses in Australia, the United States and other countries have a relatively lower degree of virus variation.","PeriodicalId":137505,"journal":{"name":"2022 Asia Conference on Algorithms, Computing and Machine Learning (CACML)","volume":"41 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123210350","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}

引用次数: 0

Review of Multi-Object Tracking Based on Deep Learning 基于深度学习的多目标跟踪研究综述

2022 Asia Conference on Algorithms, Computing and Machine Learning (CACML) Pub Date : 2022-03-01 DOI: 10.1109/CACML55074.2022.00125

Jiaxin Li, Lei Zhao, Zhaohuang Zheng, Ting Yong

引用次数: 0

Research on Data-driven Enterprise Raw Material Ordering and Optimal Transportation Scheme 数据驱动企业原材料订购及最优运输方案研究

2022 Asia Conference on Algorithms, Computing and Machine Learning (CACML) Pub Date : 2022-03-01 DOI: 10.1109/CACML55074.2022.00114

Kerun Mi, Hai-hua Gu, Jiacheng Liu, Liu Yang

{"title":"Research on Data-driven Enterprise Raw Material Ordering and Optimal Transportation Scheme","authors":"Kerun Mi, Hai-hua Gu, Jiacheng Liu, Liu Yang","doi":"10.1109/CACML55074.2022.00114","DOIUrl":"https://doi.org/10.1109/CACML55074.2022.00114","url":null,"abstract":"Ordering and transportation optimization of raw material is a classic problem in the field of optimization. Based on the analysis, evaluation and prediction of previous data, this paper fully considers the actual situation of ordering, transshipment and storage, and establishes a multistage and large-scale mixed integer linear programming model for ordering and transporting raw materials. Combined with extensive data background, select appropriate indicators, and establish supplier evaluation systems by using entropy weight-CRITIC and RTOPSIS method. GM (1,1) is used to predict the general trend, the ARIMA model is used to predict the random fluctuation items, and a grey time series prediction model is constructed to obtain the predicted values of the data of the supply and loss rate in the next cycle. The prediction result are introduced into the planning model as parameters, and the evaluation score are used to construct a satisfaction function. The final goal of mixed integer linear programming model, as well as the three goals of sorting, transferring and storing, are obtained by reduction process. Finally, this paper uses Gurobi to solve practical problems, and obtains the ordering scheme and transshipment scheme that are superior to the historical schemes in ordering cost, transshipment loss and storage cost.","PeriodicalId":137505,"journal":{"name":"2022 Asia Conference on Algorithms, Computing and Machine Learning (CACML)","volume":"141 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124456744","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}

引用次数: 0

GGAT: Knowledge Graph Embedding Model via Global Information 基于全局信息的知识图谱嵌入模型

2022 Asia Conference on Algorithms, Computing and Machine Learning (CACML) Pub Date : 2022-03-01 DOI: 10.1109/CACML55074.2022.00077

Zhe Wang, Zhongwen Guo

引用次数: 0

A quantum learning approach based on Hidden Markov Models for failure scenarios generation 基于隐马尔可夫模型的故障场景生成量子学习方法

2022 Asia Conference on Algorithms, Computing and Machine Learning (CACML) Pub Date : 2022-03-01 DOI: 10.1109/CACML55074.2022.00019

A. Zaiou, Younès Bennani, Basarab Matei, M. Hibti

引用次数: 2

COVID-19 positive cases prediction based on LSTM algorithm and its variants 基于LSTM算法及其变体的COVID-19阳性病例预测

2022 Asia Conference on Algorithms, Computing and Machine Learning (CACML) Pub Date : 2022-03-01 DOI: 10.1109/CACML55074.2022.00052

Shiqi Liu, Yuting Zhou, Xuemei Yang, Junping Yin

引用次数: 3

Algorithmic Design of Autonomous Housekeeping Robots through Imitation Learning and Model Predictive Control 基于模仿学习和模型预测控制的自主家政机器人算法设计

2022 Asia Conference on Algorithms, Computing and Machine Learning (CACML) Pub Date : 2022-03-01 DOI: 10.1109/CACML55074.2022.00024

Fangyu Zhu, Zhe Wu

引用次数: 0

End to End Multi-object Tracking Algorithm Applied to Vehicle Tracking 端到端多目标跟踪算法在车辆跟踪中的应用

2022 Asia Conference on Algorithms, Computing and Machine Learning (CACML) Pub Date : 2022-03-01 DOI: 10.1109/CACML55074.2022.00068

Wenyuan Qin, Hong Du, Xiaozheng Zhang, Xuebing Ren

{"title":"End to End Multi-object Tracking Algorithm Applied to Vehicle Tracking","authors":"Wenyuan Qin, Hong Du, Xiaozheng Zhang, Xuebing Ren","doi":"10.1109/CACML55074.2022.00068","DOIUrl":"https://doi.org/10.1109/CACML55074.2022.00068","url":null,"abstract":"At present, most of the existing multi-object tracking algorithms use the tracking-by-detection structure. On the one hand, these methods can not make full use of the intermediate features of the detector, on the other hand, the way to solve the similarity does not take into account the correlation between objects. At the same time, the existing multi-object tracking methods do not deal with the occluded object features. Based on the above problems, this paper proposes an end-to-end multi-object tracking algorithm, which uses the object deep features transmitted by the detector to directly generate the incidence matrix through the end-to-end association network; At the same time, considering the interference in occlusion, the self attention mechanism is used to enhance the features of the object. In terms of association strategy, this paper uses Hungarian matching algorithm to associate according to the association matrix. The algorithm has carried out a large number of experiments on KITTI data set, achieved 51.80% HOTA (high-order tracking accuracy) and 53.77% MOTA (multi-object tracking accuracy), and achieved considerable results compared with some existing mainstream methods.","PeriodicalId":137505,"journal":{"name":"2022 Asia Conference on Algorithms, Computing and Machine Learning (CACML)","volume":"104 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127027156","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}

引用次数: 1

Stacked Generative Adversarial Networks for Image Generation based on U-Net discriminator 基于U-Net鉴别器的图像生成堆叠生成对抗网络

2022 Asia Conference on Algorithms, Computing and Machine Learning (CACML) Pub Date : 2022-03-01 DOI: 10.1109/CACML55074.2022.00132

Wanyan Feng, Zuqiang Meng, L. Wang

{"title":"Stacked Generative Adversarial Networks for Image Generation based on U-Net discriminator","authors":"Wanyan Feng, Zuqiang Meng, L. Wang","doi":"10.1109/CACML55074.2022.00132","DOIUrl":"https://doi.org/10.1109/CACML55074.2022.00132","url":null,"abstract":"Although Generative Adversarial Networks (GANs) are powerful generative models and have shown remarkable success in various tasks recently but suffers from generating high-quality images. In this paper, we proposed a U-Net-based discriminator structure on the network structure of the two-stage Stacked Generative Adversarial Networks (StackGAN++), aiming to generate high-resolution images with actual shapes and textures. To gain more insight from limited datasets, we focused on improving the discriminator's ability to discriminate the real from the fake. The discriminator based on U-Net architecture allows providing details per-pixels and global feedback to the generator to maintain the global coherence of synthetic images and the realistic of local shape and textures. In addition, for the problem that the training effect on a small number of sample datasets is not ide-al, we further improve the quality of the generated samples by transfer learning of model parameters. Compared with the StackGAN++ baseline, experiments show that we have significantly improved the IS and FID evaluation indicators of the ImageNet subset.","PeriodicalId":137505,"journal":{"name":"2022 Asia Conference on Algorithms, Computing and Machine Learning (CACML)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130560948","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}

引用次数: 0