共查询到20条相似文献，搜索用时 187 毫秒
1.
采用在HSV彩色空间的色调累积直方图和边缘直方图分别表示颜色和形状特征,进行相似性检索,并结合综合权重调整的相关反馈技术来满足用户的检索需求。实验结果表明,此算法能获得有效的检索效果。 相似文献
2.
人脸跟踪是视频检索、视觉监控等领域中的一项关键技术,随着机器视觉技术的发展,实现人脸跟踪的算法也是多种多样。将人脸跟踪问题近似于一个meanshift跟踪问题,选取面部的颜色特征作为跟踪模板,并在MATLAB中进行了仿真,得到了良好的实验效果。 相似文献
3.
4.
5.
6.
师菊香 《内蒙古科技与经济》2011,(18):43-44
介绍了智能辅助实训系统中的机器视觉。机器视觉系统位于智能辅助实训系统的前端,是其核心组成部分。其作用是实时拍摄实训作业图像,并采集图像、转换传送到PC机内存中,PC机通过图像处理软件完成对实训操作正确、错误的判断后,将判断信息传输给辅导输出系统,从而在PC机上输出准确的实时辅导和评价信息。 相似文献
7.
8.
传统彩色图像的直方图检索技术基于RGB空间,但研究发现,RGB空间不符合人眼的视觉感应。因此本文提出的算法是在HSV空间上计算彩色图像的三维直方图,通过非均匀的划分三个分量色相H(Hue)、饱和度S(Saturation)和亮度V(Value)的采样区间,得到不同的检索结果。实验结果分析表明,在HSV空间,不同的分量在不同类型的图像中的作用是不一致的。 相似文献
9.
图像增强是数字图像的预处理,能有效地改善图像整体或局部特征。本文主要讨论了直方图均衡化处理的图像增强,并给出了相关的推导公式和算法;同时通过MATLAB实验加以验证,给出图像在处理前后的对照图像、具体算法、实验结果及直方图。实验结果表明,用直方图均衡化的算法,能有效地改善图像的对比度和灰度动态范围,使处理后的图像视觉效果得以改善。 相似文献
10.
基于直方图均衡化的医学图像增强 总被引：1，自引：0，他引：1
基于直方图均衡化的图像增强是数字图像的预处理技术,对图像整体和局部特征都能有效地进行改善。本文主要研究比较了直方图均衡化和自适应直方图均衡化的图像增强的基本原理和算法,并试验改进邻域灰度方差的取值范围。 相似文献
11.
In this paper, a novel on-line observer-based trajectory tracking strategy for leader-follower formation of multiple nonholonomic mobile robots is developed. In the proposed strategy, a leader robot follows a certain trajectory whereas a number of followers track the leader as specified by a formation protocol. Unlike other techniques in the literature, a predefined trajectory is not required, and it can be changed on-line. Moreover, this strategy aims to have a fast transient response without showing undesired overshoots. To achieve this feature, a new observer is introduced. Based on the output of that observer, a control strategy with two components is derived. The first control component is responsible for tracking the desired trajectory, whereas the second control component is used to regulate the robot to its desired steady state position. The stability of the closed loop control system is investigated. Applications of the proposed observer-based controller to different case studies are presented to illustrate the effectiveness, robustness and applicability of the developed technique. To show the superiority of proposed controller, its performance in a trajectory tracking application is compared to that of a Lyapunov-based controller. 相似文献
12.
多仿生机器鱼群体及单体控制体系结构的研究 总被引：1，自引：0，他引：1
多仿生机器鱼群体协作与控制的研究,是针对未来仿生机器鱼在水下作业、海洋考察、军事侦察等方面的实际应用而开展的相关理论和关键技术的研究。控制体系结构是其中很重要的一个研究内容,也是今后进一步开展研究工作的基础。考虑到机器鱼自主性和协作性的要求,本文给出了多仿生机器鱼群体和单体控制体系结构,并重点讨论了单体控制体系结构。该体系结构是混合分层的,具有较好的开放性,便于功能的扩展 相似文献
13.
本以Mirosot机器人足球比赛的视觉子系统为研究平台，探讨了视觉子系统的硬件构成，提出的基于颜色表的图像分割算法和目标定位算法，处理速度快，能满足系统的实时性要求。 相似文献
14.
电子政务环境下政府危机管理系统的框架研究 总被引：15，自引：0，他引：15
电子政务的快速发展对现代危机进行有效预防、快速处理和综合评估提供了有利条件。本文提出了电子政务环境下政府危机管理系统的总体框架,并详细介绍了以危机安全预警子系统、危机状态跟踪子系统、危机应急处理子系统、危机影响评估及理赔子系统四大部分为主体的框架及其主要功能。该框架可作为电子政务环境下我国建立政府危机管理系统的基础。 相似文献
15.
This paper presents a minimal-neural-networks-based design approach for the decentralized output-feedback tracking of uncertain interconnected strict-feedback nonlinear systems with unknown time-varying delayed interactions unmatched in control inputs. Compared with existing approximation-based decentralized output-feedback designs using multiple neural networks for each subsystem in lower triangular form, the main contribution of this paper is to provide a new recursive backstepping strategy for a local memoryless output-feedback controller design using only one neural network for each subsystem regardless of the order of subsystems, unmeasurable states, and unknown unmatched and delayed nonlinear interactions. In the proposed strategy, error surfaces are designed using unmeasurable states instead of measurable states and virtual controllers are regarded as intermediate signals for designing a local control law at the last step. Using Lyapunov stability theorem and the performance function technique, it is shown that all signals of the total controlled closed-loop system are bounded and the transient and steady-state performance bounds of local tracking errors can be preselected by adjusting design parameters independent of delayed interactions. 相似文献
16.
Generally speaking, the mobile robot is capable of sensing its surrounding environment, interpreting the sensed information to obtain the knowledge of its location and the environment, and planning a real-time trajectory to reach the object. In this process, the issue of obstacle avoidance is a fundamental topic to be challenged. Thus, a switching path-planning control scheme is designed without detailed environmental information, large memory size, and heavy computation burden in this study for the obstacle avoidance of a mobile robot. In this scheme, the robot can gradually approach its object according to the motion tracking mode, obstacle avoidance mode, self-rotation mode, and robot state selection designed by learning and expert rules for enhancing the tracking speed and adapting to different environments. The effectiveness of the proposed adaptive path-planning control scheme is verified by numerical simulations and experimental results of a differential-driving mobile robot under the possible occurrence of obstacle shapes. 相似文献
17.
为了解决图像检索中颜色直方图无法描述颜色空间位置信息的问题,提出了一种新的基于色彩量化的凸壳内角直方图特征.针对量化后HSV颜色空间中某一具体量化颜色值对应像素点的空间分布,首先计算包含这些像素点位置信息的凸壳,并对该凸壳的所有内角建立直方图.然后处理所有量化颜色值提取整幅图像的凸壳内角直方图特征.将该方法和一些文献中提出的方法进行的比较实验结果表明:提出的新特征可以有效地描述颜色空间位置信息,在查准率和查全率指标上均优于已有文献中的一些方法. 相似文献
18.
19.
In this paper, the trajectory tracking control problem of a six-degree of freedom (6-DOF) quadrotor unmanned aerial vehicle (UAV) with input saturation is studied. Applying a hierarchical control structure, a priori-bounded control thrust and the desired orientations are derived to stabilize the translational subsystem without singularities. By using a backstepping approach with a Nussbaum function, a priori-bounded control torque for the rotational subsystem is designed to track the desired orientations generated by the translational subsystem. With the proposed control scheme, the latent singularities in the attitude extraction process caused by saturation nonlinearities are avoided, and globally uniformly ultimately bounded (UUB) stability of the closed-loop system is achieved. The tracking error bound is further determined which can be made arbitrarily small by tuning certain control gains. Numerical simulation results are provided to show the effectiveness of the proposed control scheme. 相似文献
20.
四轮全向机器人是一个复杂的四维冗余系统,其轨迹跟踪控制性能极易受到模型不确定性及外界干扰的影响。针对这一问题,本文提出一种基于动力学模型的四轮全向机器人滑模轨迹跟踪控制方法。首先,通过输入变换将复杂的四维冗余机器人系统模型转化为三维模型,然后针对模型不确定性及外界扰动,采用滑模控制算法进行轨迹跟踪控制。仿真结果表明,该方法能够有效抑制外界干扰,同时降低模型不确定性的影响,机器人能够很好地跟踪期望轨迹,跟踪速度快,跟踪精度高。 相似文献