Sitting posture is closely related to our health. Poor sitting can cause various diseases and harm physical Current methods detect include machine vision, wearable sensors, pressure sensors. However, these have problems with respect privacy, inconvenience, cost. In this work, we proposed the use of frequency-modulated continuous wave radar (FMCW) for detecting human posture, which employs wireless signal transmission enable non-contact detection, protect reduce costs. First, range fast...

Accurately detecting falls and providing clear directions for the fall can greatly assist medical staff in promptly developing rescue plans reducing secondary injuries during transportation to hospital. In order facilitate portability protect people’s privacy, this paper presents a novel method direction motion using FMCW radar. We analyze based on correlation between different states. The range–time (RT) features Doppler–time (DT) of person from state fallen were obtained by analyzed two...

Iris image quality critically determines iris recognition performance and the metrics of images are also useful prior information for adaptive selection optimal strategy. is jointly determined by multiple factors such as focus, occlusion, off-angle, deformation, etc. So it a complex problem to assess overall score an image. This paper proposes novel framework comprehensive assessment quality. The contributions include three aspects: (i) Three approaches proposed estimate (QM) defocus, motion...

For iris recognition, it is inevitable to encounter a large portion of off-angle images in less constrained conditions. This paper proposes feature-level solution recognition which dependent on image preprocessing. Firstly, we use geometric features corneal reflections and multiclass SVM classify into five categories (i.e., frontal, right, left, up down) according the orientation region. And then feature learning method based linear programming used select most effective ordinal each...

With the development of iris recognition, many identity authentication applications began to use this inherent biometric ID. Despite breakthroughs in identification with recognition technology, one primary problem remains unsolved: presentation spoof attack. In paper, we present a novel algorithm 4DCycle-GAN for expanding image database by synthesizing fake images wearing textured contact lenses. The proposed follows Cycle-Consistent Adversarial Networks (Cycle-GAN) framework which...

The iris recognition performance is partially dependent on the relative quality variations of pairwise images. So bridging gap between and matching score images helpful to predict improve performance. This paper formulates relationship as a statistical regression problem. Firstly, number measures such focus, motion blur, illumination, off-angle, occlusions dilation are computed related feature vector And then partial least squares used establish two models intra inter from image...

Iris semantic segmentation in less-constrained scenarios is the basis of iris recognition. We propose an end-to-end trainable model for segmentation, namely Seg-Edge bilateral constraint network (SEN). The SEN uses edge map and coarse to constrain optimize mutually produce accurate results. generated from low level convolutional layers passes detailed information region by high constrains filtering scope which makes aware focusing on interesting objects. Moreover, we pruning filters...

In speech emotion recognition, most emotional corpora generally have problems such as inconsistent sample length and imbalance of categories. Considering these problems, in this paper, a variable input CRNN deep learning model based on Focal Loss is proposed for recognition anger, happiness, neutrality sadness IEMOCAP corpus. model, Firstly, variable-length strategy introduced to the spectra filled samples into CNN. Then effective part sequence preserved output by masking matrix convolution...

The cosmetic contact lenses over an iris may change original textural pattern which is the foundation for recognition, making a possible and easy-to-use presentation attack means. For practical application scenes, detection still facing unsolved problems, due to low image quality difficulty in accurately localization. In this paper, we propose novel framework called Weighted Region Network (WRN) detect lenses. WRN includes local attention Weight global classification Network. With inherent...

Autonomous ground robots autonomously are being used inthe places where it is very hazardous for human beings to reach and operate,such as nuclear power plants chemical industries. The aim of the researchpresented here develop a control system that enables such robotsnavigate with various sensors depth camera, 2D scanninglaser, 3D Lidar, GPS, IMU. controller uses current positionmeasured using on Husky A200, given waypoints thedestination. Then calculates best possible route based...

Image matching technology is the key for indoor vision robot positioning and map construction. The Scale Invariant Feature Transform(SIFT) algorithm has scale rotation invariance, but there a problem of long registration time poor real-time performance. Aiming at this problem, dual-threshold Featurnes From Accelerated Segment Test(FAST) proposed to transform feature extraction method SIFT algorithm. At same time, 128-dimensional point descriptor traditional reduced 32-dimensional, so that...

Automated detection of cervical lesion cell/clumps in cytological images is essential for computer-aided diagnosis. In this task, the shape and size appeared to vary considerably, reducing performance cell/clumps. To address issue, we propose an adaptive feature extraction network detection, called AFE-Net. Specifically, module acquire features cell/clumps, while introducing global bias mechanism average information, aiming at combining with information improve representation target model,...

Abstract Simultaneous Localization and Mapping (SLAM) is an autonomous localization technique used for mobile robots without GPS. Since relies on pre-existing maps, to use SLAM with the Robotic Operating System (ROS), a map of surroundings must first be created, controller can then initial map. The mapping procedure mostly carried out manually, human intervention. When operating person robot responsible avoiding obstacles moving different sections space create full entire environment....

DDS technology is an innovative circuit architecture. It has the advantages of low phase noise, fast frequency switching, small cubage, high resolutions. This paper presents a agile synthesizer with simple structure, anti-jamming, stray and noise.

Abstract Radar non-contact monitoring of vital signs has a broad application prospect in clinical monitoring. Aiming at the problem strong interference detection (Such as multi-target, random body motion), blind source separation (BSS) signal method based on Fast-ICA is proposed to reduce multi-target. In this algorithm, entropy used evaluate non Gaussian property, and appropriate transformation matrix selected, according statistical independence signals, signals are separated from observed...

Because different wavelet sub-bands contain image information after a RSI (Remote Sensing Image) being transformed by wavelet, compression method based on adaptive threshold has been proposed. According to entropy theory, the amount of sub-band contains can be expressed its entropy, and each was set depending entropy. To reduce much computation relation between other statistic value for analyzed. And we found average absolute had clear steady with By curve fitting, mathematical expression...

Many professions rely on disguised voice techniques. In contrast to the conventional linear prediction-based modification technique, we propose a algorithm with smoothing filtering improve effect by changed pitch period. This paper is based prediction. The cepstrum method used obtain and change period of speech residual signal, applying process obtained period, synthesis realized using prediction filter achieve same rate. results show that synthesized signal tends be more clear natural...

A low power consumption image acquisition system with the core controller of CPLD, based on LVDS interface and CMOS digital sensor was designed here. We analyzed work timing ship DS90C124 MT9V011, simulated VHDL code relevant drive circuit, did an actual debug operation. The experimental results show that this has advantages low-power high reliability, it can be used in a variety processing systems.