搜索资源列表
tsp(2)
- 本程序实现了多城市的tsp问题(即旅行商问题)的遗传算法-the process of multi-city tsp (ie, the traveling salesman problem) Genetic Algorithm
EMRPC
- 路由选择过程的c算法程序 与DSR 只是简单地使用最短路径不同,EMRPC 使用了更复杂的路由选择机制以实现多径路由和能量均衡。-process of routing algorithm c procedures and DSR is simply to use the shortest path different. EMRPC use of a more complex routing mechanisms to achieve multi-path routing and energy
Agentbasedbilateralalgorithm.
- 双边多议题协商是一个复杂的动态交互过程。解决Agent 在对环境和对方信息不全知的情况下通过协商达成一致并最大化自身效用是非常重要的。 -Bilateral multi-issue consultations is a complex process of dynamic interaction. Solve the Agent in the environment and the other incomplete information to know where to reach agr
bp
- bp算法实现,运行正常。人工神经网络中BP算法实现了多层反馈式学习过程-bp algorithm, operating normally. BP artificial neural network algorithms in multi-layer feedback-based learning process
duomubiaoyichuan
- 多目标遗传算法应用于电火花机床参数优化,优化过程简单,计算结果准确,收敛速度明显加快-Multi-objective genetic algorithm optimization applied to EDM, the optimization process is simple, accurate calculations, significantly speeding up the convergence rate
mock_handl_demo
- mock_handl_demo mock聚类程序,多目标进化算法应用于聚类的程序,经过调试通过,效果不错-mock_handl_demo mock clustering process, multi-objective evolutionary algorithm is applied to clustering process, after debugging through, good results
完整的多机器人路径规划程序
- 一个完整的多机器人路径规划程序,几代研究生都在用-A complete multi-robot path planning process, several generations of graduate students are using
GA-pareto
- 遗传算法的多目标优化算例,动态显示优化过程pareto前端的分布情况。-Genetic algorithm is a multi-objective optimization example, dynamic display of pareto optimal process the distribution of the front end
Multi-Agent-Particle-Swarm-Algorithm
- 结合多智能体的学习、协调策略及粒子群算法,提出了一种基于多智能体粒子群优化的配电网络重构方法。该方法采用粒子群算法的拓扑结构来构建多智能体的体系结构,在多智能体系统中,每一个粒子作为一个智能体,通过与邻域的智能体竞争、合作。能够更快、更精确地收敛到全局最优解。粒子的更新规则减少了算法不可行解的产生,提高了算法效率。实验结果表明,该方法具有很高的搜索效率和寻优性能。-Combining the study of multi-agent technology,coordinating strateg
FaceRecognitionBased-OnDeepLearning
- 本文运用深度神经网络的方法克服姿态变量和图像分辨率的影响,提出了一种多姿态的人脸超分辨识别算法并在实验数据集上获得了良好的性能表现。另外本文利用深度信念网络探索正面人脸图像和侧面人脸图像的映射,方法放松了深度信念网络的输入也输出之间绝对相等,而只是保证其高层含义上的相等。实验表明了使用深度信念网络可以学习到侧面人脸图像到正面人脸图像的一个全局映射,但是丢失了个体细节差异。本文还提出了基于深度网络保持姿态邻域进行姿态分类的方法,在学习过程中,我们保证了同一个姿态下的人脸图像应该与同一姿态下的多张图
classify
- This tutorial booklet introduces you to the TNTmips procedures for automatic classification of multispectral and multi-temporal imagery. The process includes “unsupervised” methods
Masaike_Detection
- 视频在解码过程中,有时候需要检测是否含有马赛克帧,及时去除以提高视频质量。这里利用多线程,实现了视频解码过程中马赛克实时检测。-Video in the decoding process, and sometimes need to detect whether it contains mosaic frame, timely removal to improve the video quality. Here the use of multi-threaded, to achieve the
我的编码
- 多目标函数的遗传算法求解过程的编码,伪代码,轮盘赌算法(Multi objective function genetic algorithm to solve the process of coding, pseudo code, roulette algorithm)
多目标优化进化算法
- 按照外文经典文献的步骤和流程,在matlab环境下编写了多目标优化进化算法程序。在外文经典文献中,有该算法的子函数开源代码,程序运行速度快、精度高。文件由主程序和子函数构成,共7个.m文件。(According to the steps and process of foreign literature, a multi-objective optimization evolutionary algorithm program is written in Matlab environment.
4人工势场法路径规划程序
- 多机器人系统的群集编队控制理论仿真程序,采用人工势场法,使多个机器人在向目标点运动过程中能保持队形,并能适应环境约束(The simulation program of cluster formation control for multi robot system is based on artificial potential field method, which enables multiple robots to maintain formation in the process of
fast-rcnn-master
- Fast R-CNN是在R-CNN的基础上进行的改进,大致框架是一致的。总体而言,Fast R-CNN相对于R-CNN而言,主要提出了三个改进策略: 1. 提出了RoIPooling,避免了对提取的region proposals进行缩放到224x224,然后经过pre-trained CNN进行检测的步骤,加速了整个网络的learning与inference过程,这个是巨大的改进,并且RoIPooling是可导的,因此使得整个网络可以实现end-to-end learning,这个可以认为是
