摘要 据不完全统计,在 2014 年,我国因交通事故死亡人数约为 35000 人,其中疲劳驾驶无疑是诱发交通事故的主要原因之一,因此致力于疲劳驾驶预警系统的研究,在避免交通事故方面具有深刻的现实意义。 本文在了解国内外相关研究的基础上,为了满足疲劳驾驶检测非接触、实时性和全天候的要求,从硬件和软件两方面设计出一套基于 DSP 的疲劳驾驶检测的设计方案: (1)在硬件方面,选取了以 TMS320DM642 芯片为核心的各模块芯片,并给出了电压转换模块、红外光源模块、视频解码电模块、声光报警模块、时钟模块、FLASH 模块、SDRAM 模块的具体电路连接图,并对各个电路逐一分析。 (2)在软件算法方面,本文放弃了传统的先定位人脸,再定位人眼的思路,利用亮瞳效应,选取了差分图像的方法快速定位人眼,节省运算时间,这部分的图像处理通过MATLAB仿真。在完成图像处理的各个环节后,提取了 PERCLOS和眨眼频度作为疲劳参数判断疲劳状态。 本文在设计整套系统的同时,验证了整个系统的正确性和可行性,为实际应用打下基础。 35724
毕业论文关键词: DSP 人眼检测定位 红外差分 PERCLOS
Abstract According to incomplete statistics, in 2014, fatigue driving in our country is undoubtedly one of the main causes of induced traffic accident with traffic accident death toll at about 35000 people.Committed to the research of fatigue driving warning system, in terms of traffic accident avoidance has profound practical significance. On the basis of the understanding of domestic and foreign related research, in order to satisfy the fatigue driving detection of non-contact, real-time and all-weather requirements, this paper designs a fatigue detection based on DSP design scheme from two aspects of hardware and software: (1) in the aspect of hardware, selected the TMS320DM642 chip as the core of each module, and presents a voltage conversion module, the infrared light source module, video decoding module, sound and light alarm module, clock module, the circuit connection diagram FLASH module, SDRAM module, and one by one for each circuit analysis. (2) in terms of software algorithm, this paper gave up the traditional positioning face first, then locate the thinking of the human eye, use bright pupil effect, selecting the difference image method to quickly locate the human eye, saves the operation time, this part of the image processing by the MATLAB simulation. On completion of each link of the image processing, to extract the PERCLOS and blink frequency asfatigue parameters determine the fatigue status. In the design of the system at the same time, this paper verified the correctness and feasibility of the whole system, and lay the foundation for practical applications.
Keywords: DSP human eye detection positioning PERCLOS infrared difference
目录
摘要I
AbstractII
1绪论1
11疲劳驾驶研究的背景与意义1
12疲劳检测的常用方法2
13疲劳驾驶检测国内外研究现状2
131国外研究现状2
132国内研究现状3
14论文结构安排4
2系统设计思路6
21常规人脸检测方法6
22亮瞳效应7
23硬件整体设计框架8
24软件整体流程图8
3硬件系统的设计10
31DSP芯片的选取10
32摄像头的选取12
33光源电路12
34视频采集电路13
331视频解码芯片TVP5150PBS介绍13
332视频解码电路14
35电源电路15
36时钟电路17
37SDRAM电路19
38Flash电路20
39JTAG接口及声光报警电路21
310系统复位电路22
4软件系统的设计23
41人眼定位23
411图像平滑23
412图像差分处理25
- 上一篇:高层建筑下多级水发电系统仿真分析+CAD图纸+Matlab源代码
- 下一篇:LabVIEW输气管道泄漏监测系统设计+源代码
-
-
-
-
-
-
-
g-C3N4光催化剂的制备和光催化性能研究
现代简约美式风格在室内家装中的运用
江苏省某高中学生体质现状的调查研究
NFC协议物理层的软件实现+文献综述
高警觉工作人群的元情绪...
巴金《激流三部曲》高觉新的悲剧命运
上市公司股权结构对经营绩效的影响研究
C++最短路径算法研究和程序设计
浅析中国古代宗法制度
中国传统元素在游戏角色...