Earticle

For several years, keyboard and mouse have been used as the main interacting devices between users and computer games, but they are becoming outdated. Gesture-based human- computer interaction systems are becoming more popular owing to the emergence of virtual reality and augmented reality technologies. Therefore research on these systems has attracted a significant attention. The researches focus on designing the interactive interfaces between users and computers. Human-computer interaction is an important factor in computer games because it affects not only the experience of the users, but also the design of the entire game. In this research, we develop an particle filter-based face tracking method using color distributions as features, for the purpose of applying to gesture-based human-computer interaction systems for computer games. The experimental results proved the efficiency of particle filter and color features in face tracking, showing its potential in designing human-computer interactive games.

Target tracking algorithm is a key technical problem in the computer version field, and the existing algorithms are mainly based on the stationary camera platform. Therefore, a real-time tracking algorithm for moving target based on dynamic monitoring video is proposed in this article. Firstly, the feature points are screened according to the influence of the moving target on the light streams; secondly, the light stream method is adopted to estimate the global motion of the camera; finally, the motion equation of the particle filter is modified according to the global motion estimation result and meanwhile the color histogram is selected as the characteristic model of the target, thus to track the moving target. The experiment result shows that the particle filter algorithm can rapidly and accurately track the moving target during camera motion process to meet the real-time requirement, and has excellent robustness.

In view of the problem of particle degradation and tracking accuracy in the standard particle filter tracking target algorithm, a new improved particle filter algorithm called Iterated Cubature Kalman Particle Filter (ICKPF) is proposed in this paper. The new ICKPF algorithm is based on the Markov Chain Monte Carlo (MCMC), and the cubature rule based on numerical integration method is used to calculate the mean and covariance, which generates the proposal distribution for the particle filter. The current measurements are integrated into the proposal distribution. Therefore, degree of approximation to the system posterior density is improved. Simulation results show that the estimation error of the ICKPF-MCMC algorithm is much better than other algorithms.

Urban road traffic state identification is a key link to realize the intelligent transportation based on the Internet of Vehicles, and accurately positioning vehicles is the foundation to realize the traffic state identification. Aiming at the problem that GPS has signal blind area in positioning vehicles, a vehicle positioning algorithm based on particle filter was proposed, it could improve the traditional algorithm on degradation and large amount of calculations; Based on vehicle positioning, an urban road traffic state identification algorithm based on fuzzy discrimination was proposed, it could comprehensively consider multiple factors’ influence on traffic state. The experiment results show that the improved particle filter algorithm’s mean squared error has increased about 55.437% compared with GPS method, and the traffic state identification algorithm can accurately identify the traffic state of the study area, it can prove that the urban road traffic state identification algorithm based on particle filter and fuzzy discrimination is feasible and effective.

In this article, we present an improved resampling algorithm for particle filtering, which is based on partial resampling and residual resampling. This algorithm provides an approach to selectively carry out hierarchical resampling operations on three sets of particles divided by large, medium and small weights, and especially to do skip resampling for partial small particles with an adaptive interval M. Simulation results verify that the proposed algorithm could reduce the depletion problem, maintain a good diversity of particles and improve the accuracy of PF performance.

The paper proposes Gaussian mixture particle probability hypothesis density filter(PHD) algorithm ,which can effectively solve the problem that the object number is changing or unknown, based on particle PHD filter. This algorithm calculates the object number and state by recursive procedure, avoiding the uncertainty of target state estimation caused by particle sampling and clustering. Gaussian mixture particle is introduced to effectively maintain the multi-modal distribution of each target，reducing the complexity of calculation.

The accurate state-of-charge (SOC) estimation for lithium-ion battery (LIB) cells and packs plays an important role in fulfilling efficient battery management. However, some estimation errors of LIB states and SOC are often encountered when using the conventional battery model and filter algorithms. Thus, this paper explores a SOC estimation method of LIB based on an improved partnership for a new generation of vehicle (PNGV) model and particle filter (PF) algorithm. First, a second-order PNGV model is constructed, and the effect of different SOC interval changes on battery model parameters is considered. Then, the improved PNGV model-based PF algorithm is employed to realize the real-time estimation of battery SOC. Finally, the proposed SOC estimation method is validated under constant current and dynamic urban dynamometer driving schedule (UDDS) conditions. Results show that the improved PNGV model of LIB integration with the PF algorithm can considerably enhance the accuracy of SOC estimation. Compared to the conventional PNGV model-based PF algorithm, the mean absolute error and root mean squared error of SOC estimation errors are, respectively, reduced by approximately 32.7% and 32.5% under the 0.5C rate discharge and UDDS conditions.

Most of the indoor positioning algorithms based on magnetic data mainly focus on reducing the accumulated error of the odometry data, such as signals produced by the inertial sensors. However, in most cases such as positioning by using smartphones in the indoor environment, those approaches seem unfeasible due to the absence of the inertial sensors. Thus, in this paper, we try to study a positioning algorithm exclusively based on the magnetic data. We refer to some thinking from the steps of Particle Filter and conduct an experiment to verify the application of the new algorithm. Besides, we use the variance of the result of the previous step to decrease the area to be matched in the next step, intending to improve the accuracy of the results. The result of the experiment shows that the new algorithm has a high probability to match with accuracy less than 2 meters in a 24 meters by 2.6 meters corridor.

컴퓨터 비젼은 지난 30년간 빠른 속도로 발전해왔다. 1970년대 초에는 인공지능분야가 굉장히 다양하게 연구되었는데, 그중 Condensation 알고리즘으로 알려진 블레이크 교수의 연구를 시작으로 많은 연구자들이 이 알고리즘을 연구하기 시작했다. 초기 단계에서 이 알고리즘은 파티클 필터라고 알려져 있고, 근래 신호처리 분야에서 두각을 나타내고 있다 하지만 파티클 필터는 내부적인 계산과정 때문에 FrameGrabber가 필요하게 되어 하드웨어 없이는 실시간 처리가 힘들고, 1분에 30프레임의 동영상 규격보다 훨씬 더 많은 50프레임을 사용하고 있다. 이에 본 논문에서는 동영상처리기법에 있어서 DirectX 기반의 제안된 프레임 컨트롤이라는 기법을 사용하여, 변화하는 동영상에 능동적으로 반응하는 하드웨어 없는 프레임 레이트(Framerate) 컨트롤 엔진을 사용하여, 실시간 인물 추적 알고리즘을 구현하려 한다.

본 논문에서는 CT 영상에서 스네이크 알고리즘과 파티클 필터를 이용한 뼈 영역 추출 방법을 제안한다. 스네이크 알고리즘을 이용하여 뼈 외곽선을 추출하고, 이 외곽선을 따라서 파티클 필터를 움직여 가면서 뼈 영역을 추출한다. 뼈 외곽선 주위에 다른 뼈가 근접해 있는 경우에 스네이크 알고리즘이 뼈 외곽선을 완벽하게 추출하지 못할 수 있는데, 이때에 파티클 필터가 이러한 오류를 보완해가면서 뼈 영역을 추출한다. 본 논문에서 제안한 방법을 기존의 형태학적 처리를 이용한 방법과 비교한 결과, 추출하고자 하는 뼈 영역 주위에 다른 뼈가 근접해 있지 않은 경우에는 비슷한 결과를 보였지만, 다른 뼈가 근접해 있는 경우에는 본 논문에서 제안한 방법의 정확도가 높음을 알 수 있었다.

In this paper, we propose a bone region extraction method using a snake algorithm and a particle filter in CT image. We extract the bone outline using the snake algorithm, and extract the bone area by moving the particle filter along this outline. If other bones are in close proximity to the bone outline, the snake algorithm may not be able to extract the bone outline completely. At this time, the particle filter extracts the bone area while compensating for the error. In this paper, we compared the proposed method with the conventional morphological processing method. The result is similar when other bones are not close to the bone area to be extracted. However, if other bones are close to each other, The accuracy of the proposed method is higher than the conventional morphological processing method.

자동차의 안전성 향상에 대한 연구는 오랜 기간 다양한 분야에서 진행되고 있다. 이 시스템은 단일 카메라를 이용하여 차선을 감지함으로써 차선 침범을 방지한다. 시스템은 파티클 필터를 이용해 도로 모델 파라미터를 조정하고 두 개의 detector가 도로 모델의 일치도를 계산한다. Detector는 차선의 모양과 색이라는 대표적인 특징을 이용한다. 파티클 필터를 전 프레임에서 사용한 모델 파라미터를 이용하여 신속한 처리를 한다.