2019 12th International Congress on Image and Signal Processing, BioMedical Engineering and Informatics (CISP-BMEI)最新文献

{"title":"Human Pose Estimation via Multi-Scale Intermediate Supervision Convolution Network","authors":"Zhuohao Mai, Xiao Hu, Shao-Hu Peng, Yunshan Wei","doi":"10.1109/CISP-BMEI48845.2019.8965911","DOIUrl":"https://doi.org/10.1109/CISP-BMEI48845.2019.8965911","url":null,"abstract":"Human pose estimation is a technique for locating the key points of a human in the picture and video. It can be used in the fields of human-computer interaction, motion recognition, Character behavior analysis and so on. At present, the supervision information is a series of single-scale heat map without matching the multi-scale key points of ground truth while training the CNN model. This type of supervision makes it more likely that the predicted key points will deviate from the real position. To improve the prediction accuracy, this paper proposes a novel model named human pose estimation via multi-scale intermediate supervision convolution network. Heatmaps are generated on three scale which are determined by the standard deviation of 2D Gaussian distribution. Residual network model is composed of three stage using ResNet50 as the backbone network. Each stage includes a ResNet50 and three deconvolution layers. The output from ResNet50 in three stages corresponds to the heat maps annotation of large, medium and small sizes respectively, and the intermediate supervision is realized twice in the output of the first and second stages. In the test phase, the output from the last stage is used for calculating the final key point's coordinates by non-maximum suppression. To demonstrate effectiveness of our network, two benchmark datasets are used for training and testing: the key point detection subset of COCO dataset and MPII Human Pose dataset. The test result of PCK@0.1 reached 37.2% on the MPII validation dataset, which is 2.1% higher than other methods. The results of the mAP test on the COCO validation dataset reached 75.5%, an increase of 1.2% compared with other methods. The results indicate that the multi-scale relay supervised convolutional network model proposed in this paper can reduce the influence of the non-correspondence between the size of key points and the size of heatmap ground truth in human pose estimation, thus improving the accuracy and achieving better performance when the evaluation criteria are stricter.","PeriodicalId":257666,"journal":{"name":"2019 12th International Congress on Image and Signal Processing, BioMedical Engineering and Informatics (CISP-BMEI)","volume":"17 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129280929","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}

{"title":"A Fast Eyelash Detection Algorithm Based on Morphological Operation","authors":"Shuqin Gao, Min Han, Deqiang Wang, Meng Wang","doi":"10.1109/CISP-BMEI48845.2019.8965795","DOIUrl":"https://doi.org/10.1109/CISP-BMEI48845.2019.8965795","url":null,"abstract":"Iris recognition is regarded as a promising biometric recognition technology. However, iris recognition schemes suffer from the interference from eyelashes. Intuitively, it is a crucial issue to detect the eyelash precisely in the preprocessing stage of iris recognition. In this paper, a novel eyelash detection algorithm in iris image based on morphological operation is proposed. Given an iris image, the effective search area of the eyelash is determined based on the position of the pupil; Then, morphological closing operation is performed on the effective search area of the eyelash and thereby a difference image is got. After that, the difference image is segmented by using an optimal threshold which is obtained by using maximum interclass variance method and thus candidate eyelash pixels are determined. Finally, based on the candidate eyelash pixels, a two-stage eyelash detection process is carried out to figure out the real eyelash pixels. Experimental results suggest that the proposed algorithm can achieve good effect of eyelash detection, and compared with the previous methods, the detection speed is greatly increased, it meets the requirements of speed and accuracy during the image preprocessing stage.","PeriodicalId":257666,"journal":{"name":"2019 12th International Congress on Image and Signal Processing, BioMedical Engineering and Informatics (CISP-BMEI)","volume":"139 2","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114107397","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}

{"title":"Design and Implementation of the Performance Test of Our Self-developed Bowel Sound Recorder","authors":"Guojing Wang, Weidong Wang, Hongyun Liu","doi":"10.1109/CISP-BMEI48845.2019.8965959","DOIUrl":"https://doi.org/10.1109/CISP-BMEI48845.2019.8965959","url":null,"abstract":"Bowel sound is one of the important physiological signals of human body and the research of detecting and analyzing is increasing and deepening. However, there are few special electronic devices for the bowel sound, and there is almost no performance test for the corresponding electronic devices. As for the special bowel sound recorder developed by ourselves, this paper carried out tests in the special audiometry room, and used the first-level sound level meter as the standard instrument to determine the performance parameters of our bowel sound recorder. The test results showed that (1) the upper range value of the bowel sound recorder can reach 100dB; (2) the trend of the test results between the bowel sound recorder and sound level meter is consistent; (3) the maximum deviation between the bowel sound recorder and the sound level meter is 10dB, and this result provides a theoretical basis for subsequent calibration. Through the quantitative testing, the validity and reliability of bowel sound recorder are determined, which lays a hardware foundation for the subsequent collection of clinical data and is of great clinical significance for the effective acquisition of bowel sounds.","PeriodicalId":257666,"journal":{"name":"2019 12th International Congress on Image and Signal Processing, BioMedical Engineering and Informatics (CISP-BMEI)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124337669","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}

{"title":"Sensorimotor Control of Reach-to-Pinch Kinematics in Individuals with Alzheimer's Disease","authors":"Leitong Lin, Ke Li, Na Wei","doi":"10.1109/CISP-BMEI48845.2019.8966065","DOIUrl":"https://doi.org/10.1109/CISP-BMEI48845.2019.8966065","url":null,"abstract":"Alzheimer's disease (AD) is a chronic neurodegenerative disease that could reduce the sensory and motor function. This study aimed to investigate the effects of AD on sensorimotor control of hand kinematics during a reach-to-pinch a target. Eight patients with AD and eight healthy volunteers participated in the experiment. Retro-reflective markers were attached to the fingertips of the thumb and index finger, hand and wrist. The kinematic data were recorded using a 3D motion capture system during reach to pinch a custom designed instrument with and without visual guidance. The mean error of contact position in the x direction showed that the mean error increased in AD compared with the controls (AD = 52.2 mm; HC = 12.7 mm, p < 0.05). The AD patients also demonstrated increased variability of the distances between the contact points and the target without visual guidance (AD = 25.9 mm; HC = 11.2 mm, p < 0.05). However, for the movement duration, there was no significant difference between the AD and the controls. These results revealed that the AD affects the capacity of sensorimotor control of reach-to-pinch kinematics by decreasing the pinch accuracy but increasing the pinch variability. This finding could be applied in clinical diagnosis and evaluation of AD.","PeriodicalId":257666,"journal":{"name":"2019 12th International Congress on Image and Signal Processing, BioMedical Engineering and Informatics (CISP-BMEI)","volume":"21 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122270348","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}

{"title":"Footwear Design for Hallux Valgus: A Virtual Prototyping with Finite Element Modeling","authors":"Zong-Yuan Ying, Bing-Shan Hu, Xin Ma, Wenming Chen","doi":"10.1109/CISP-BMEI48845.2019.8965988","DOIUrl":"https://doi.org/10.1109/CISP-BMEI48845.2019.8965988","url":null,"abstract":"In this paper, we use virtual prototype with finite element modeling to customize footwear design for hallux valgus foot with the purpose of offloading plantar peak pressure from the local metatarsal head. The footwear consists of two material layers to achieve deformation patterns during foot-insole contact interactions (i.e., soft tissue squeezes the insole FE model). The deformed model of the footwear is exported and a metatarsal pad is added in different placement. From the results of simulation, the metatarsal pad and the deformation synergistically achieve the purpose of decompression, and an optimal design could be found by virtually varying with the position of metatarsal pad.","PeriodicalId":257666,"journal":{"name":"2019 12th International Congress on Image and Signal Processing, BioMedical Engineering and Informatics (CISP-BMEI)","volume":"5 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131915781","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}

{"title":"Design of QPSK Sequences for M-Ary Spread Spectrum OFDM Systems","authors":"Guixin Xuan, Zhenyu Zhang, Chenrun Liu","doi":"10.1109/CISP-BMEI48845.2019.8965673","DOIUrl":"https://doi.org/10.1109/CISP-BMEI48845.2019.8965673","url":null,"abstract":"As an efficient tradeoff method between data rate and communication quality, M-ary spread spectrum (MSS) orthogonal frequency division multiplexing (OFDM) system may provide a flexible choice for different communication requirements. However, conventional BPSK sequences with single channel mode limits high-speed transmission to a great extent. In order to obtain higher data rate, a design method of QPSK sequences for MSS-OFDM systems is proposed in this paper. The provided QPSK sequences are constructed on the basis of complementary orthogonal sequence set which is simultaneously mapped to both of in-phase channel and quadrature channel. As a result, the increase of twice data rate may be achieved at the cost of peak-to-average power ratio growth from 3dB to 6dB.","PeriodicalId":257666,"journal":{"name":"2019 12th International Congress on Image and Signal Processing, BioMedical Engineering and Informatics (CISP-BMEI)","volume":"8 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131946906","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}

{"title":"Image Inpainting Exploiting Tensor Train and Total Variation","authors":"Shuli Ma, Huiqian Du, Jiayun Hu, Xinyi Wen, Wenbo Mei","doi":"10.1109/CISP-BMEI48845.2019.8965787","DOIUrl":"https://doi.org/10.1109/CISP-BMEI48845.2019.8965787","url":null,"abstract":"In this paper, we propose a novel approach to RGB image inpainting, which recovers missing entries of image by using low rank tensor completion. The approach is based on recently proposed tensor train (TT) decomposition, which is used to effectively enforce the low rankness of the image. In addition, our approach exploits the local smooth priors of visual data by incorporating the 2D total variation. Ket augmentation (KA) scheme is used to permute the image to a high order tensor, and then low rankness of balanced KA-TT matrices and total variation (TV) norm constraints are applied to recover the missing entries of the image. In order to reduce the computational complexity, in the proposed approach, nuclear norm is replaced by minimum Frobenius norm of two factorization matrices, which reduces the time for singular value decomposition (SVD). Lastly, in order to solve the proposed model, the efficient alternating direction method of multipliers (ADMM) is developed. The results of image inpainting experiments demonstrate the significantly superior performance of our approach.","PeriodicalId":257666,"journal":{"name":"2019 12th International Congress on Image and Signal Processing, BioMedical Engineering and Informatics (CISP-BMEI)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131699040","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}

{"title":"Preliminary Study of Optimization of Mri-compatible Ultrasonic Transducer Matching Layer Composites with Comsol","authors":"Xinping Ren, Wenjie Liu, Zhiqiang Su, Hao Wu, S. Qiao, Huaxin Lu, G. Shen","doi":"10.1109/CISP-BMEI48845.2019.8965701","DOIUrl":"https://doi.org/10.1109/CISP-BMEI48845.2019.8965701","url":null,"abstract":"Phase controlled high intensity focused ultrasound (p-hifu) is electronic phase controlled focusing.By adjusting the excitation voltage phase and amplitude of each array element, a variety of sound field modes, such as single-focus movement and multi-focus, can be generated. Magnetic resonance imaging (MRI) can reflect tissue details in nondestructive conditions and measure the temperature of tissue in vivo by analyzing the proton resonance frequency signals. The combination of p-hifu technology and MRI technology improves the accuracy and safety of treatment, but the combination of the two brings about non-negligible electromagnetic interference problems. Most EMC problems can be solved by electromagnetic shielding. Based on previous studies, this paper intends to study a novel electromagnetic shielding material Cu-Epoxy for ultrasonic transducers.The biggest innovation of this material is its dual functions- electromagnetic shielding and ultrasonic matching layer.We conduct preliminary simulation with Comsol software to verify its feasibility. Through the simulation of electromagnetic field and piezoelectric effect, we find that this material is feasible in theoretical design. The biggest advantage of this simulation model is that it can be used for theoretical design of composite materials and explore the optimization of their performance. In further research, we can develop the model from the simplest two-dimensional axisymmetric model to a three-dimensional model, making it closer to the real situation.","PeriodicalId":257666,"journal":{"name":"2019 12th International Congress on Image and Signal Processing, BioMedical Engineering and Informatics (CISP-BMEI)","volume":"10 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130690158","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}

{"title":"Dual Discriminator Generative Adversarial Network for Single Image Super-Resolution","authors":"Piaoyi Yuan, Yaping Zhang","doi":"10.1109/CISP-BMEI48845.2019.8965727","DOIUrl":"https://doi.org/10.1109/CISP-BMEI48845.2019.8965727","url":null,"abstract":"Single image super-resolution(SISR) is to reconstruct a high resolution(HR) image from a single low resolution(LR) image. In this paper, with generative adversarial networks(GAN) model as the basic component, we build a dual discriminator super-resolution reconstruction network(DDSRRN) to improve the quality of image super-resolution reconstruction. By adding another discriminator based on GAN, we combine the Kullback-Leibler(KL) with reverse KL divergence to make a unified objective function to train the two discriminators, and by using the complementary statistical characteristics of these two divergences, the prediction density is effectively dispersed in multi-mode, which can avoid collapse of the network model during the reconstruction process and improve the stability of model training. We build the content loss function using the Charbonnier loss and use the intermediate features information of the two discriminators to build the perceptual loss function and style loss function. The experimental results show that the proposed method has sharp edges and rich details in subjective vision, and obtains better subjective visual evaluation and objective quantitative evaluation.","PeriodicalId":257666,"journal":{"name":"2019 12th International Congress on Image and Signal Processing, BioMedical Engineering and Informatics (CISP-BMEI)","volume":"231 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131092537","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}

{"title":"A Wireless Flexible Pressure Sensor for Human Motion Detection","authors":"Rong Huang, Yu Zhang, Xinjian Chen, Baoqing Nie","doi":"10.1109/CISP-BMEI48845.2019.8965659","DOIUrl":"https://doi.org/10.1109/CISP-BMEI48845.2019.8965659","url":null,"abstract":"This paper presents a flexible wireless pressure sensor with a high sensitivity and quality factor which exhibits a great potential application in tactile sensing, health monitoring and smart electronics. Our flexible wireless pressure sensor is built on a three-layer structure: the bottom layer is a planar antenna consisting of an inductor and an inner-digit capacitor, which are patterned on two opposite-side of polyimide substrate, connected in serious; The middle layer is a layer of compressible flexible sponge with a thickness of 4.4mm; The top layer is a ferrite film with a thin layer of adhesive. Under external pressure, the flexible sponge layer is mechanically deformed, leading to the distance between the ferrite film and the antenna layer decreases. Consequently, the magnetic permeability increases, resulting in an increase in inductance, and further resonant frequency decreases. This study sets a special space to research the impact of the environment on the sensor (from indoor temperature to high temperature and from indoor humidity to high humidity). It turns out that environmental factors have little effect on the sensor, reflecting the environmental stability of the sensor. The cyclic and bending experiments have also been performed. Our sensor exhibits excellent stability when pressed and released periodically or bent repeatedly. Finally, we demonstrate the utility by attaching the sensor to an elastic leg straps to measure the compression force of the strap under different leg motions.","PeriodicalId":257666,"journal":{"name":"2019 12th International Congress on Image and Signal Processing, BioMedical Engineering and Informatics (CISP-BMEI)","volume":"78 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132277043","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}