This article is aimed at designing a human identity recognition algorithm based on face image, which will be used in indoor environment. As the working environment is set as indoor environment, the camera will not be ...
This article is aimed at designing a human identity recognition algorithm based on face image, which will be used in indoor environment. As the working environment is set as indoor environment, the camera will not be affected much by illumination variation. The key is how to detect human face and handle the variation of facial pose. This article divides the whole recognition process into 4 parts: image pre-processing, face detection, face alignment, feature extraction and comparison. Face detection and feature extraction are the core functions and both realized by deep learning. The process of the whole algorithm can be described as following: images after pre-processing are fed to face detection network to get the locations of face and face landmarks. Then face alignment will be conducted. Finally, deep features of face will be extracted and compared. The unique features of this algorithm are its good performance of handling the variation of facial pose and its clear framework which allows the whole method can be easily adjusted and upgraded.
The technology of the detection for vehicle and driver is a popular spot in these ten years. In particular, the driver detection is still a troubled question in the study of public security. In our paper, an algorithm...
The technology of the detection for vehicle and driver is a popular spot in these ten years. In particular, the driver detection is still a troubled question in the study of public security. In our paper, an algorithm based on YOLOv3 and support vector machine (SVM) is proposed for realizing the detection of vehicles on highway, as well as the detection and binary classification of people in the vehicles, so as to achieve the purpose of distinguishing drivers and passengers and form a one-to-one correspondence between vehicles and drivers. The effectiveness of the algorithm is verified under various complicated highway conditions. Compared with other advanced vehicle and driver detection technologies, the model has a good performance and is robust to road blocking, different attitudes and extreme lighting.
This paper will introduce the design of the multidimensional Taylor network(MTN) optimal control in the flight control of aircraft. The MTN optimal control, which combines the classical architecture o
ISBN:
(纸本)9781467389808
This paper will introduce the design of the multidimensional Taylor network(MTN) optimal control in the flight control of aircraft. The MTN optimal control, which combines the classical architecture o
Moving vehicle detection based on video processing has been widely used in intelligent transportation system recently. However there are also many problems, such as dynamic background, ghost region, and shadow of movi...
Moving vehicle detection based on video processing has been widely used in intelligent transportation system recently. However there are also many problems, such as dynamic background, ghost region, and shadow of moving objects. This paper proposes an improved ViBe object detection algorithm. First, an accurate background image is obtained by using the multi-frame averaging method, and then the background model is initialized by this accurate background image, thus effectively reducing the generation of ghost region. Whenever there is no moving object for a fixed number of consecutive frames in the video, this frame is updated to the background image. Conservative update strategy and foreground point counting method are adopted to update the background and reduce the impact of dynamic background on the foreground detection. Next, the foreground image detected by improved ViBe algorithm is input into the shadow elimination method proposed in this paper. Shadows in foreground pixels are detected in RGB color space, and then the pixels determined as shadows are eliminated. Finally, accurate moving vehicles are obtained. Our algorithm can effectively eliminate the shadows of moving vehicles, quickly adapt to background changes and illumination mutation, and get accurate moving objects, which is helpful for vehicle contour extraction and subsequent image processing.
With the development of the electricity market, competition has been introduced in the generation side. It is the overall development trend of the electricity market reformulation to optimize the allocation of differe...
详细信息
This paper considers the problem of global stabilization for a class of high-order nonlinear systems whose states cannot be measured accurately and the uncertain parameters occur in the powers of the measurement funct...
详细信息
In the practical application,Brushless DC motor(BLDCM) faces various disturbances including parametric uncertainties,load disturbance and unmodeled *** disturbances can be divided into two types of periodic disturbanc...
详细信息
ISBN:
(纸本)9781467397155
In the practical application,Brushless DC motor(BLDCM) faces various disturbances including parametric uncertainties,load disturbance and unmodeled *** disturbances can be divided into two types of periodic disturbances(sinusoidal/cosinoidal) and aperiodic(slowly-varying or constant) *** ripples can be seen as the result of a plurality of periodic interference signals of different frequencies acting on the *** the paper we propose a composite control scheme combining integral sliding mode control(ISMC) based on disturbance observer(DO) embedded with the internal model principle to improve the speed performance of *** disturbance observer,such as extended state observer(ESO) can only asymptotically estimate slowly-varying or constant disturbances and is not good at estimating periodic *** we combine internal model of disturbance into disturbance observe for high ***,the estimates are introduced in the feedforward compensation,and a composite speed controller is *** last,experimental comparisons with conventional proportional-integral(PI) and ISMC+ESO,are given to validate the effectiveness of the proposed method.
Since proportional-integral-derivative (PID) controllers absolutely dominate the controlengineering, numbers of different control structures and theories have been developed to enhance the effciency of PID controller...
详细信息
In the copper electrolytic refining process, the short-circuit failure among the electrodes usually exists. When it occurs, the temperature distribution is often abnormal, accompanied by local or overall overheating. ...
详细信息
In the copper electrolytic refining process, the short-circuit failure among the electrodes usually exists. When it occurs, the temperature distribution is often abnormal, accompanied by local or overall overheating. This paper proposes a novel processing method for infrared temperature images to effectively detect the short-circuit fault and locate the accurate malfunctioning spots. The design procedure contains contrast enhancement, filter denoising, edge detection, image segmentation, high temperature electrode positioning. Finally, experimental results show that the proposed method can guarantee real-time processing of the infrared image of the electrolytic cell and accurately mark the malfunctioning position of electrodes with high temperature fault.
This paper studies the similar formation algorithm of multi-agent systems with biased measurement errors. When relative position measurements contain biased errors, the existing similar formation algorithm is not mean...
详细信息
ISBN:
(纸本)9781509015740;9781509015733
This paper studies the similar formation algorithm of multi-agent systems with biased measurement errors. When relative position measurements contain biased errors, the existing similar formation algorithm is not mean reachable in general. This paper proposes a modified similar formation algorithm with which agents can overcome biased measurement errors online and reach any desired generic formation shape in the expectation sense. The formation error is globally uniformly bounded in the mean square sense.
暂无评论