Proceedings of the 2022 11th International Conference on Computing and Pattern Recognition最新文献_第7页

Anomaly Handwritten Text Detection for Automatic Descriptive Answer Evaluation 用于自动描述答案评估的异常手写文本检测

Proceedings of the 2022 11th International Conference on Computing and Pattern Recognition Pub Date : 2022-11-17 DOI: 10.1145/3581807.3581855

Nilanjana Chatterjee, Palaiahnaakote Shivakumara, U. Pal, Tong Lu, Yue Lu

{"title":"Anomaly Handwritten Text Detection for Automatic Descriptive Answer Evaluation","authors":"Nilanjana Chatterjee, Palaiahnaakote Shivakumara, U. Pal, Tong Lu, Yue Lu","doi":"10.1145/3581807.3581855","DOIUrl":"https://doi.org/10.1145/3581807.3581855","url":null,"abstract":"Although there are advanced technologies for character recognition, automatic descriptive answer evaluation is an open challenge for the document image analysis community due to large diversified handwritten text and answers to the question. This paper presents a novel method for detecting anomaly handwritten text in the responses written by the students to the questions. The method is proposed based on the fact that when the students are confident in answering questions, the students usually write answers legibly and neatly while they are not confident, they write sloppy writing which may not be easy for the reader to understand. To detect such anomaly handwritten text, we explore a new combination of Fourier transform and deep learning model for detecting edges. This result preserves the structure of handwritten text. For extracting features for classification of anomaly text and normal text, the proposed method studies the behavior of writing style, especially the variation at ascenders and descenders. Therefore, the proposed work draws principal axis which is invariant to rotation, scaling and some extent to distortion for the edge images. With respect to principal axis, the proposed method draws medial axis using uppermost and lowermost points. The distance between the medial axis and principal axis points are considered as feature vector. Further, the feature vector is passed to Artificial Neural Network for classification of anomaly text. The proposed method is evaluated by testing on our own dataset, standard dataset of gender identification (IAM) and handwritten forgery detection dataset (ACPR 2019). The results on different datasets show that the proposed work outperforms the existing methods.","PeriodicalId":292813,"journal":{"name":"Proceedings of the 2022 11th International Conference on Computing and Pattern Recognition","volume":"49 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-11-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124609506","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 Lightweight Brain Tumor Segmentation Network Based on 3D Inverted Residual Modules 基于三维倒立残差模块的轻量级脑肿瘤分割网络

Proceedings of the 2022 11th International Conference on Computing and Pattern Recognition Pub Date : 2022-11-17 DOI: 10.1145/3581807.3581829

Yuchao Liu, X. Du, Da-han Wang, Shunzhi Zhu

引用次数: 0

Immersive AR Merged with MI-BCI Hand Function Rehabilitation Training System for Stroke Patients 沉浸式AR与脑卒中患者MI-BCI手功能康复训练系统的融合

Proceedings of the 2022 11th International Conference on Computing and Pattern Recognition Pub Date : 2022-11-17 DOI: 10.1145/3581807.3581851

Yiyang Qin, Banghua Yang, Dongze Li

{"title":"Immersive AR Merged with MI-BCI Hand Function Rehabilitation Training System for Stroke Patients","authors":"Yiyang Qin, Banghua Yang, Dongze Li","doi":"10.1145/3581807.3581851","DOIUrl":"https://doi.org/10.1145/3581807.3581851","url":null,"abstract":"Strokes can cause neurological damage to the patient, which leads to hand dysfunction. Traditional methods of hand function rehabilitation, such as electrical stimulation and therapist-dependent movement therapy, are ineffective due to the brain's lack of direct involvement in the motor nervous system. To improve the rehabilitation efficacy, we design a rehabilitation system based on motor imagery brain-computer interface (MI-BCI) and augmented reality (AR) for hand function rehabilitation of stroke patients. It includes two-class motor imagery tasks: left-hand fist and right-hand fist based on AR. Motor imagery electroencephalogram (MI-EEG) is acquired from 10 subjects and decoded by using an algorithm module encapsulated in the master system. It reaches an average accuracy of 76.4% and is eventually fed back to patients through rehabilitation peripherals. In addition, the master system provides an interactive interface with features to design treatment tasks, manage patient information and monitor patient status. The system realizes an immersive rehabilitation experience that promotes the reconstruction of the central nervous system and provides a new approach for stroke patients to recover.","PeriodicalId":292813,"journal":{"name":"Proceedings of the 2022 11th International Conference on Computing and Pattern Recognition","volume":"36 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-11-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128167568","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 Improved Regularization Method Based on Signal Waveform Measurement 基于信号波形测量的改进正则化方法研究

Proceedings of the 2022 11th International Conference on Computing and Pattern Recognition Pub Date : 2022-11-17 DOI: 10.1145/3581807.3581862

Jiangmiao Zhu, Wenshu Yu, Kejia Zhao, Zhaotong Wan

引用次数: 0

The Classification Method of EEG Motor Imagery Based on INFO-LSSVM 基于INFO-LSSVM的脑电运动图像分类方法

Proceedings of the 2022 11th International Conference on Computing and Pattern Recognition Pub Date : 2022-11-17 DOI: 10.1145/3581807.3581876

Xinrong Wang, Abdelkader Nasreddine Belkacem, Penghai Li, Zufeng Zhang, Jun Liang, Dongdong Du, Chao Chen

{"title":"The Classification Method of EEG Motor Imagery Based on INFO-LSSVM","authors":"Xinrong Wang, Abdelkader Nasreddine Belkacem, Penghai Li, Zufeng Zhang, Jun Liang, Dongdong Du, Chao Chen","doi":"10.1145/3581807.3581876","DOIUrl":"https://doi.org/10.1145/3581807.3581876","url":null,"abstract":"For the current situation that the classification accuracy of EEG motor image data is not high in the BCI system, a vector weighted average algorithm optimization algorithm is proposed, and the optimized least squares support vector machine algorithm is proposed to classify the EEG motor image data. A motor imagination EEG experimental paradigm was designed and compared with the unoptimized LSSVM and three other typical classification methods on the same dataset. The experimental data were band-pass filtered by the fourth-order Butterworth filter of 0.5-30Hz, and the electrical interference was removed by independent component analysis. The HHT features obtained by empirical mode decomposition (EMD) and Hilbert Yellow transform (HHT) in the time-frequency domain were input into INFO-LSSVM for classification. Compared with dense feature fusion convolutional neural network (DFFN), Restricted Boltzmann machine optimized support vector Machine classifier (RBM-SVM) and public space pattern based artificial Neural network (CSP-ANN) classification algorithm, the highest classification accuracy of the proposed algorithm is 92.13%, and the average accuracy is 90.325%. It can be seen that compared with the existing algorithms with higher performance, the proposed algorithm effectively improves the classification accuracy and can better classify and identify EEG signals, which provides a new optimization idea for people's EEG signal classification.","PeriodicalId":292813,"journal":{"name":"Proceedings of the 2022 11th International Conference on Computing and Pattern Recognition","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-11-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130277348","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

Deep Features Based IDS Alarm False Positive Elimination Algorithm 基于深度特征的IDS报警误报消除算法

Proceedings of the 2022 11th International Conference on Computing and Pattern Recognition Pub Date : 2022-11-17 DOI: 10.1145/3581807.3581890

Jie Cheng, Jingchu Wang, Ang Xia, Lu Teng, Jianyi Liu

引用次数: 0

A 3D U-Net-Based Approach for Intracranial Aneurysm Detection 基于三维u - net的颅内动脉瘤检测方法

Proceedings of the 2022 11th International Conference on Computing and Pattern Recognition Pub Date : 2022-11-17 DOI: 10.1145/3581807.3581816

Tianyu Zhu, Xinfeng Zhang, Xiaomin Liu, Xiangsheng Li, Maoshen Jia, Xiaoxia Chang, Yuan Meng

{"title":"A 3D U-Net-Based Approach for Intracranial Aneurysm Detection","authors":"Tianyu Zhu, Xinfeng Zhang, Xiaomin Liu, Xiangsheng Li, Maoshen Jia, Xiaoxia Chang, Yuan Meng","doi":"10.1145/3581807.3581816","DOIUrl":"https://doi.org/10.1145/3581807.3581816","url":null,"abstract":"Intracranial aneurysm refers to a neoplastic protrusion of the arterial wall caused by a localized abnormal enlargement of the cerebral artery lumen. In clinical practice, patients in the early stage of onset generally have no obvious symptoms, which is very easy to miss diagnosis. In medicine, methods such as MRA, CTA and DSA can be used to display the images of blood vessels. Among them, magnetic resonance angiography (MRA) has the advantages of low cost and small damage to the human body. Which can display the images of blood vessels in the brain. The data set used herein was based on images provided by a three-dimensional time-of-flight magnetic resonance angiography system. The main contributions of this paper are as follows: (1) We improved a classic 3D U-Net model with the combination of attention gate, residual connection, and the changes of size. Which achieved automatic segmentation of aneurysms in MRA. In the detection of aneurysms with mean diameters of 6.10mm and 7.69mm, the sensitivity was 83.4% and 86.4% respectively. (2) On the basis of this sensitivity, we achieved a low false positive rate which was 0.36 FPs/case and 0.34 FPs/case respectively. CCS CONCEPTS • Computing methodologies∼Computer graphics∼Image manipulation∼Image processing","PeriodicalId":292813,"journal":{"name":"Proceedings of the 2022 11th International Conference on Computing and Pattern Recognition","volume":"170 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-11-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123468471","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 Novel Chinese Resume Named Entity Recognition Model Based on Lexical Enhancement 一种基于词汇增强的中文简历命名实体识别模型

Proceedings of the 2022 11th International Conference on Computing and Pattern Recognition Pub Date : 2022-11-17 DOI: 10.1145/3581807.3581856

Jinshang Luo, Ying Liu, Mengshu Hou

{"title":"A Novel Chinese Resume Named Entity Recognition Model Based on Lexical Enhancement","authors":"Jinshang Luo, Ying Liu, Mengshu Hou","doi":"10.1145/3581807.3581856","DOIUrl":"https://doi.org/10.1145/3581807.3581856","url":null,"abstract":"The resume's popularity on the Internet has greatly increased with the development of the communication form. It is a concern of researchers to analyze the resumes of job applicants using the Named Entity Recognition (NER) method. The difficulty of Chinese Resume NER rests with word segmentation ambiguity and domain knowledge complexity. To tackle the issue, a novel lexical enhancement Long Short-Term Memory (LSTM) model with the average encoding strategy (LEAE-LSTM) is proposed. First, through the pre-trained models, the representations of characters and words are encoded separately. The lexical features with complementary information are introduced for the character sequence by matching the lexicon. Furthermore, to improve contextual awareness, the multi-metadata embeddings are combined as the input of the LSTM layer. The sentence's implicit correlations are picked up by the self-attention mechanism. Experiments on the benchmark resume dataset demonstrate that LEAE-LSTM surpasses other state-of-the-art methods. For the Chinese resume dataset, LEAE-LSTM gains a 1.8% improvement in F1 score over the baseline model Lattice LSTM.","PeriodicalId":292813,"journal":{"name":"Proceedings of the 2022 11th International Conference on Computing and Pattern Recognition","volume":"5 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-11-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122260498","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

RT-Net: Plant Phenotype Semantic Segmentation Network Based on Advanced Deep Learning Framework RT-Net:基于高级深度学习框架的植物表型语义分割网络

Proceedings of the 2022 11th International Conference on Computing and Pattern Recognition Pub Date : 2022-11-17 DOI: 10.1145/3581807.3581831

Chengui Fu, Wenbiao Xie, Yin Jin, Kai Zhao, Qiuming Liu, He Xiao

{"title":"RT-Net: Plant Phenotype Semantic Segmentation Network Based on Advanced Deep Learning Framework","authors":"Chengui Fu, Wenbiao Xie, Yin Jin, Kai Zhao, Qiuming Liu, He Xiao","doi":"10.1145/3581807.3581831","DOIUrl":"https://doi.org/10.1145/3581807.3581831","url":null,"abstract":"Quantitatively deriving plant phenotypes from plant images in a non-contact manner is a very challenging task that relies heavily on the accurate segmentation of plant images. Previous methods mainly used the U-Net network structure and attention mechanism to obtain the corresponding plant phenotype segmentation results. However, the U-Net structure and attention mechanism are relatively outdated, and its method can only achieve a Dice score of 98.47% on the open source dataset, which is still insufficient for the recent plant phenotype segmentation task and needs to be further improved for more detailed research. Therefore, in view of the low segmentation performance of existing plant phenotype semantic segmentation models, this paper proposes a semantic segmentation network RT-Net based on an advanced deep learning framework. Specifically, the network mainly adopts the encoder-decoder network structure of deeplabv3+, and the encoding part of the network adopts the more efficient RepVGG as the backbone network for local feature extraction. At the same time, compared with the traditional Atrous Spatial Pyramid Pooling (ASPP), this paper designs the (Atrous Spatial Pyramid Pooling Based Transformer)ASPPBT module to extract more global feature information through a global adaptive method to obtain denser plant phenotypes. The decoding part performs feature fusion on the output of the encoding part, and then uses upsampling to restore the scale, and finally obtains the semantic segmentation result. The experimental results show that the proposed network has achieved a Dice score of 99.33% on the Arabidopsis plant dataset released by the CVPPP14 competition, and has better segmentation ability compared with other advanced plant field segmentation algorithms","PeriodicalId":292813,"journal":{"name":"Proceedings of the 2022 11th International Conference on Computing and Pattern Recognition","volume":"60 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-11-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127968142","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

LEPD-Net: A Lightweight Efficient Network with Pyramid Dilated Convolution for Seed Sorting leped - net:一种用于种子排序的轻量级高效金字塔扩展卷积网络

Proceedings of the 2022 11th International Conference on Computing and Pattern Recognition Pub Date : 2022-11-17 DOI: 10.1145/3581807.3581888

Weijie Li, Pingsun Wei, Jun Sun, Xiaoting Xiao, X. Mu, Zhenghui Hu

{"title":"LEPD-Net: A Lightweight Efficient Network with Pyramid Dilated Convolution for Seed Sorting","authors":"Weijie Li, Pingsun Wei, Jun Sun, Xiaoting Xiao, X. Mu, Zhenghui Hu","doi":"10.1145/3581807.3581888","DOIUrl":"https://doi.org/10.1145/3581807.3581888","url":null,"abstract":"To achieve long-term economic growth, competitiveness, and sustainability, speed and accuracy are the key requirements when it comes to seed purity sorting. However, current seed sorting methods suffer from large number of model parameters and computational complexity, make it a great challenge to deploy them in real-time applications, especially on devices with limited resources. To issue above problems, in this paper, a lightweight efficient network with pyramid dilated convolution, namely LEPD-Net, is proposed for seed sorting. First, a residual spatial pyramid module (RSPM) is elaborately designed, which uses dilated convolution with different dilation rates to enlarge the structural characteristics of the receptive field and effectively extracts multi-scale features. Then the depth-wise separable convolution to reduce the amount of model parameters and the computational complexity. In addition, to further improve the performance, a novel lightweight coordinate attention module is introduced, which uses the local cross-channel interaction to obtain the attention value of each channel and strengthen the network's ability to learn seed key features. Finally, the seed sorting task is completed through the learned features. Experimental results show that our proposed method achieves an accuracy of 96.00% and 97.25% on the Maize dataset and Red Kidney Bean dataset, respectively. The number of parameters is only 0.26M, which is far less than state-of-the-art networks (e.g., MobileNetv2, Shufflenetv2, and PPLC-Net).","PeriodicalId":292813,"journal":{"name":"Proceedings of the 2022 11th International Conference on Computing and Pattern Recognition","volume":"4 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-11-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127987984","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