2021 5th Asian Conference on Artificial Intelligence Technology (ACAIT)最新文献_第4页

Research on English Grammar Error Correction Technology Based on BLSTM Sequence Annotation 基于BLSTM序列标注的英语语法纠错技术研究

2021 5th Asian Conference on Artificial Intelligence Technology (ACAIT) Pub Date : 2021-10-29 DOI: 10.1109/acait53529.2021.9731256

Yanzhai Shi

引用次数: 1

Recognition of parasite eggs in microscopic medical images based on YOLOv5 基于YOLOv5的显微医学图像中寄生虫卵的识别

2021 5th Asian Conference on Artificial Intelligence Technology (ACAIT) Pub Date : 2021-10-29 DOI: 10.1109/acait53529.2021.9731120

Yibo Huo, Jing Zhang, Xiaohui Du, Xiangzhou Wang, Juanxiu Liu, Lin Liu

{"title":"Recognition of parasite eggs in microscopic medical images based on YOLOv5","authors":"Yibo Huo, Jing Zhang, Xiaohui Du, Xiangzhou Wang, Juanxiu Liu, Lin Liu","doi":"10.1109/acait53529.2021.9731120","DOIUrl":"https://doi.org/10.1109/acait53529.2021.9731120","url":null,"abstract":"Parasitosis is a disease caused by parasites invading the human body. Because of the different species and parasitic sites, it causes different pathological changes and clinical manifestations, and also causes other complications, which is harmful to human health. In clinical medicine, the diagnosis of parasitic diseases is mostly through etiological diagnosis, that is, through the detection of whether there are parasitic eggs in human feces. The diagnosis and treatment of parasitic diseases is a very important part of clinical medicine. At present, the recognition and classification of parasite eggs in human fecal microscopic images are mainly based on manual processing and machine learning, which are inefficient and easily affected by subjective factors, while machine learning can not deal with complex and changeable fecal environment. Here, an automatic recognition algorithm based on YOLOv5 for parasite eggs in fecal microscopic medical images is proposed. Experimental results show that the average accuracy of the model is 0.994 in our test set. In addition, the calculation time of each human fecal microscopic image under GPU is less than 25 ms, and the algorithm has higher accuracy and faster speed than the traditional machine learning algorithm. As such, it will help advance the etiological diagnosis of parasitic diseases and the development of therapeutic drugs.","PeriodicalId":173633,"journal":{"name":"2021 5th Asian Conference on Artificial Intelligence Technology (ACAIT)","volume":"31 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-10-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130250736","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: 4

Application of RBF Neural Network in the Construction of Intelligent Predictive Model of Public Building Energy Consumption RBF神经网络在公共建筑能耗智能预测模型构建中的应用

2021 5th Asian Conference on Artificial Intelligence Technology (ACAIT) Pub Date : 2021-10-29 DOI: 10.1109/acait53529.2021.9731231

Xing Song, Yanqing Yang

引用次数: 0

Robotic Autonomous Grasping Technique: A Survey 机器人自主抓取技术:综述

2021 5th Asian Conference on Artificial Intelligence Technology (ACAIT) Pub Date : 2021-10-29 DOI: 10.1109/acait53529.2021.9731320

Lili Wang, Zhen Zhang, Jianhua Su, Qipeng Gu

引用次数: 1

POC: Periodical Orthogonal Center Loss For Open Set Classification 开集分类的周期正交中心损失

2021 5th Asian Conference on Artificial Intelligence Technology (ACAIT) Pub Date : 2021-10-29 DOI: 10.1109/acait53529.2021.9731132

Yusheng Pu, Ruonan Liu, Qian Chen, Dongyue Chen, Wenlong Yu, Di Cao

引用次数: 0

Spatial Geometric Constraints based Iterative Clustering Algorithm 基于空间几何约束的迭代聚类算法

2021 5th Asian Conference on Artificial Intelligence Technology (ACAIT) Pub Date : 2021-10-29 DOI: 10.1109/acait53529.2021.9731164

Hang Yu, X. Yin, Rui Zhang, Chenyang Li, Haoran Jiang

{"title":"Spatial Geometric Constraints based Iterative Clustering Algorithm","authors":"Hang Yu, X. Yin, Rui Zhang, Chenyang Li, Haoran Jiang","doi":"10.1109/acait53529.2021.9731164","DOIUrl":"https://doi.org/10.1109/acait53529.2021.9731164","url":null,"abstract":"In visual measurement technology, the precise extraction of spatial feature points’ centroids is crucial to ensure the measurement accuracy of the visual system based on feature point imaging. This task is particularly challenging because of light spot blurring, gray influence, random shapes, gray impacting, and modeling limitations. The current extraction methods for light spot center can only extract single spot center at each time. As far as for multiple light spots in image, extracting their centers must be manually one by one. In this study, a spatial geometric constraints based on iterative clustering algorithm (SGCICA) is proposed and automatically extract multiple light spots’ centers through a K-means algorithm practice. Considering clustering algorithms can easily obtain multiple cluster centers in feature space, we introduce the information in image space into clustering algorithms from two aspects: (1) the pixel coordinate is adopted as the features for clustering algorithm to obtain the multiple light spots’ centers; (2) the spatial orders and geometric constraints among the spots are defined in the objective function of clustering algorithms to ensure the accurate extraction of actual LED targets. SGCICA operating in clustering feature space can effectively and naturally manage the information from image space. The spatial geometric constraints can improve the precision and the robustness of the clustering results. In experiments, the noise resistance and extraction precision of the proposed method are evaluated using synthetic and real data and compared with the existing light spot centers extraction methods and clustering algorithms. Both qualitative and quantitative measures indicate that the precision of the extracted light spot centers by SGCICA is kept within 0.04 pixels and satisfy the known order and spatial geometric constraints.","PeriodicalId":173633,"journal":{"name":"2021 5th Asian Conference on Artificial Intelligence Technology (ACAIT)","volume":"54 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-10-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126648919","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

A Differential Evolution Based Self-Adaptive Multi-Task Evolutionary Algorithm 基于差分进化的自适应多任务进化算法

2021 5th Asian Conference on Artificial Intelligence Technology (ACAIT) Pub Date : 2021-10-29 DOI: 10.1109/acait53529.2021.9731139

Jing J. Liang, Leiyu Zhang, Kunjie Yu, B. Qu, C. Yue, Kangjia Qiao

{"title":"A Differential Evolution Based Self-Adaptive Multi-Task Evolutionary Algorithm","authors":"Jing J. Liang, Leiyu Zhang, Kunjie Yu, B. Qu, C. Yue, Kangjia Qiao","doi":"10.1109/acait53529.2021.9731139","DOIUrl":"https://doi.org/10.1109/acait53529.2021.9731139","url":null,"abstract":"This paper proposes a novel self-adaptive evolutionary multi-task optimization algorithm based on differential evolution (SMTDE) for solving multiple different optimization problems or tasks simultaneously. The algorithm arranges a specific population and three differential strategies for each task. Among the three strategies, one is the transfer strategy and the others are non-transfer strategies. The transfer strategy is mainly responsible for utilizing the information of other tasks, and the two non-transfer strategies are responsible for accelerating convergence and improving the diversity of intra-task, respectively. Based on strategies, a self-adaptive mechanism is proposed to adjust the selection probabilities of the three strategies to reduce the harm of negative transfer and balance the diversity and convergence within the population. The experiment is conducted on a single-objective multi-task test suite. The experiment results show that SMTDE can find better solutions with a higher convergence rate in comparison with several competitive evolutionary multi-task optimization algorithms.","PeriodicalId":173633,"journal":{"name":"2021 5th Asian Conference on Artificial Intelligence Technology (ACAIT)","volume":"36 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-10-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127079406","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

Self-Supervised Continuous Meta-Learning for Few-shot Image Classification 基于自监督连续元学习的少镜头图像分类

2021 5th Asian Conference on Artificial Intelligence Technology (ACAIT) Pub Date : 2021-10-29 DOI: 10.1109/acait53529.2021.9731334

Suyan He, Yingjian Li

{"title":"Self-Supervised Continuous Meta-Learning for Few-shot Image Classification","authors":"Suyan He, Yingjian Li","doi":"10.1109/acait53529.2021.9731334","DOIUrl":"https://doi.org/10.1109/acait53529.2021.9731334","url":null,"abstract":"Few-shot classification aims to adapt the knowledge learned from base classes with sufficient data to new classes with limited data, where meta-learning methods are usually leveraged for this challenging task. However, most existing algorithms suffer from insufficient representation and testing bias issues, accordingly failing to exploit useful semantic information while being prone to cause the gap of classification accuracy between training classes and testing classes. To this end, we propose the Self-Supervised Continuous Meta-Learning (SS-CML) framework to simultaneously handle the mentioned problems, which consists of two key modules. i.e., Self-Supervised Embedding network and Self-Supervised GNN. Specifically, Self-Supervised Embedding network can extract informative semantic information from training images so that the learned prototype are more representative for the classification task. Moreover, Self-Supervised GNN learn reactions between nodes without true labels, which can improve the reliability of knowledge prior to classify images of new classes, thereby reducing the excessive dependence of training classes and alleviating the testing bias issue. Furthermore, these two modules are jointly leveraged in our SS-CML to generalize the prior knowledge to novel classes. Extensive experimental results on MiniImageNet and TieredImageNet show up the effectiveness of both self-supervised branches which boost classification performance.","PeriodicalId":173633,"journal":{"name":"2021 5th Asian Conference on Artificial Intelligence Technology (ACAIT)","volume":"15 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-10-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125523992","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

Research on Leakage Monitoring of Metal Oxide Surge Arrester Based on Hybrid Particle Swarm Algorithm 基于混合粒子群算法的金属氧化物避雷器泄漏监测研究

2021 5th Asian Conference on Artificial Intelligence Technology (ACAIT) Pub Date : 2021-10-29 DOI: 10.1109/acait53529.2021.9731291

Kai Zhang, Yongyan Xu

{"title":"Research on Leakage Monitoring of Metal Oxide Surge Arrester Based on Hybrid Particle Swarm Algorithm","authors":"Kai Zhang, Yongyan Xu","doi":"10.1109/acait53529.2021.9731291","DOIUrl":"https://doi.org/10.1109/acait53529.2021.9731291","url":null,"abstract":"The leakage monitoring of the metal oxide arrester(MOA) is completed in the process of extracting the resistive component of the leakage current of the arrester, that is, the leakage monitoring of the arrester is realized by analyzing the operation of the resistive component of current. It is shown in relevant studies and literature that in the process of using metal oxide arrester, resistive component of current changes due to the aging of metal oxides. Therefore, the problem appears to be the leakage of the arrester. In view of this, aiming at the problem that previous intelligent algorithms are difficult to monitor the leakage of MOA accurately, this paper proposes an online monitoring algorithm based on hybrid particle swarm optimization by combining the classical particle swarm optimization algorithm and the nonlinear classification method. That is, after constructing the objective function of MOA, the characteristic parameters C and α which can effectively reflect the operation of MOA are solved. And then the current resistive component of the leakage current function equation is extracted to monitor the leakage of the metal oxide arrester. The research results show that in this study, the characteristic parameters C and α solved by the hybrid particle swarm algorithm are 502.19 and 24.9786 respectively, and the corresponding mean errors are 0.438% and 0.086% respectively. At the same time, the resistive current curve obtained by the hybrid particle swarm optimization algorithm is closer to the actual situation than that obtained by the particle swarm optimization algorithm. Thus, it can improve the accuracy of MOA Leakage Monitoring effectively.","PeriodicalId":173633,"journal":{"name":"2021 5th Asian Conference on Artificial Intelligence Technology (ACAIT)","volume":"21 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-10-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133210013","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

Construction and application of biological visual nerve computing model in robot 生物视觉神经计算模型在机器人中的构建与应用

2021 5th Asian Conference on Artificial Intelligence Technology (ACAIT) Pub Date : 2021-10-29 DOI: 10.1109/acait53529.2021.9731313

Naigong Yu, Hejie Yu, Tong Qiu, Jia Lin

{"title":"Construction and application of biological visual nerve computing model in robot","authors":"Naigong Yu, Hejie Yu, Tong Qiu, Jia Lin","doi":"10.1109/acait53529.2021.9731313","DOIUrl":"https://doi.org/10.1109/acait53529.2021.9731313","url":null,"abstract":"Biological vision is very effective and accurate in scene classification and recognition. Based on this, this paper proposes a biological visual neural computing model based on the anatomical structure of rat brain, which is characterized by: constructing a biological visual scene memory model (visual word bag), imitating the biological brain’s storage of environmental scene information, and calculating the similarity between the current scene information and the visual template; designing and constructing the object details located in the lateral entorhinal cortex and the peripheral olfactory cortex Cell discharge model. Experimental results show that the proposed model can effectively extract image features and generate visual word bag model based on image features. Compared with ratslam scan line strength model, the retrieval time of this model is greatly shortened; The object cell discharge model with image similarity information as input can show similar expression of discharge rate as physiological research, which verifies the effectiveness and efficiency of the proposed model. The research results lay a foundation for the research of robot environment cognition method based on brain cognitive mechanism.","PeriodicalId":173633,"journal":{"name":"2021 5th Asian Conference on Artificial Intelligence Technology (ACAIT)","volume":"10 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-10-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132115428","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