【摘要】：在刑事技術(shù)中,剪切工具痕跡是其研究的重點(diǎn)方向之一。工具痕跡是盜竊、搶劫、殺人等許多案件中最常出現(xiàn)的一種痕跡,對(duì)于認(rèn)定案件性質(zhì),確定作案工具,證實(shí)犯罪嫌疑人具有重要意義。據(jù)不完全統(tǒng)計(jì),70%以上的刑事案件現(xiàn)場(chǎng)有工具痕跡,有些地區(qū)的工具痕跡出現(xiàn)比例達(dá)80%左右。而且工具痕跡還具有不易破壞、難以偽裝、出現(xiàn)率高,鑒定價(jià)值好的特點(diǎn),這些優(yōu)勢(shì)都是其它類(lèi)型痕跡難以比擬的。因而對(duì)工具痕跡的提取分析進(jìn)行深入研究,具有重大的實(shí)際意義。本文在痕跡信號(hào)的采集上面,采用基于LabVIEW設(shè)計(jì)的工具痕跡激光檢測(cè)裝置。通過(guò)該裝置可以有效的采集到工具痕跡上面的痕跡信號(hào)。在采集信號(hào)的過(guò)程中由于反光以及裝置等不可控因素的存在,所以?huà)呙璨杉玫降男盘?hào)中會(huì)有異常數(shù)據(jù)以及噪聲干擾的存在。在對(duì)異常數(shù)據(jù)處理上面,重點(diǎn)研究了運(yùn)用K-Means算法對(duì)信號(hào)中的異常數(shù)據(jù)進(jìn)行修復(fù),并通過(guò)試驗(yàn)仿真來(lái)驗(yàn)證該算法對(duì)于剪切工具痕跡激光檢測(cè)信號(hào)中的異常數(shù)據(jù)具有很好的修復(fù)效果。在對(duì)信號(hào)降噪處理上面,重點(diǎn)研究了通過(guò)LOWESS(局部加權(quán)回歸散點(diǎn)平滑法)算法對(duì)數(shù)據(jù)進(jìn)行平滑處理,通過(guò)該算法能夠最大程度的消除掃描數(shù)據(jù)中的噪聲。然后通過(guò)試驗(yàn)仿真來(lái)驗(yàn)證該算法的有效性。在相似度比對(duì)上面,首先通過(guò)對(duì)平滑后的信號(hào)進(jìn)行特征信號(hào)的提取,對(duì)提取得到的特征信號(hào)進(jìn)行特征向量處理。將信號(hào)間的比對(duì)轉(zhuǎn)化成空間距離的計(jì)算。最后利用動(dòng)態(tài)規(guī)劃進(jìn)行逐個(gè)相似度匹配,得到最終相似度的大小,進(jìn)而判斷出剪切工具。在理論研究與試驗(yàn)仿真的基礎(chǔ)上,通過(guò)工具痕跡激光檢測(cè)裝置對(duì)痕跡信號(hào)進(jìn)行采集。然后軟件實(shí)現(xiàn)和試驗(yàn)分析測(cè)試相結(jié)合,來(lái)對(duì)本文所提出來(lái)的算法進(jìn)行驗(yàn)證分析,進(jìn)而判斷出該算法的有效性和正確性。
[Abstract]:In the criminal technique, the trace of the shear tool is one of the key directions of its research. The tool mark is one of the most frequently occurring marks in many cases such as theft, robbery, and killing. It is of great significance to identify the nature of the case, determine the case and confirm the criminal suspect. According to incomplete statistics, more than 70% of the criminal cases have tool marks in the field, and some areas have some 80% of the tool marks. And the tool trace has the characteristics of being difficult to be damaged, difficult to camouflage, high in appearance rate and good in identification value, and all of these advantages are difficult to compare with other types of trace. Therefore, it is of great practical significance to study the extraction and analysis of the tool marks. In this paper, on the acquisition of trace signal, a tool-trace laser detection device based on LabVIEW is used. The device can effectively collect the trace signal on the tool trace. In the process of acquiring the signal, due to the existence of non-limiting factors such as the reflection and the device, the presence of the abnormal data and the noise interference in the signal obtained by the scanning acquisition can be present. In this paper, the method of K-Means algorithm is used to repair the abnormal data in the signal, and the test simulation is used to verify that the algorithm has good repair effect on the abnormal data in the laser detection signal of the shear tool. In this paper, we focus on the smoothing of the data through the LOWESS (local weighted regression point smoothing method) algorithm, which can eliminate the noise in the scan data to a maximum extent. And then the validity of the algorithm is verified by the test simulation. The method comprises the following steps of: firstly, carrying out characteristic signal extraction on the smoothed signal at a similarity ratio, and carrying out feature vector processing on the extracted feature signal. The ratio of the signals to the space distance is calculated. And finally, the dynamic programming is used for matching the similarity degree one by one, so that the size of the final similarity is obtained, and then the shearing tool is judged. On the basis of the theoretical research and the test simulation, the trace signal is collected by means of the tool trace laser detection device. And then the software implementation and the test analysis test are combined to carry out the verification and analysis on the algorithm presented in the paper, so as to judge the validity and the correctness of the algorithm.
【學(xué)位授予單位】：昆明理工大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2017
【分類(lèi)號(hào)】：D918.91

【相似文獻(xiàn)】

相關(guān)期刊論文 前10條

1 申金;談痕跡技術(shù)的發(fā)展[J];鐵道部鄭州公安管理干部學(xué)院學(xué)報(bào);2000年04期

2 鐘東榮;;工、槍、特痕跡觀察實(shí)驗(yàn)比較法教學(xué)探索[J];中國(guó)科技信息;2004年24期

3 楊琳,佟德鑫;對(duì)工具痕跡分類(lèi)方法的探討[J];中國(guó)人民公安大學(xué)學(xué)報(bào)(自然科學(xué)版);2005年01期

4 張宗亮;;評(píng)《中國(guó)現(xiàn)代痕跡學(xué)論綱——以建構(gòu)體系為視閾》[J];江蘇警官學(xué)院學(xué)報(bào);2013年03期

5 班茂森;;中國(guó)現(xiàn)代痕跡學(xué)理論研究的突破與亮點(diǎn)——《中國(guó)現(xiàn)代痕跡學(xué)論綱》評(píng)介[J];山東警察學(xué)院學(xué)報(bào);2013年04期

6 劉容;;痕跡學(xué)理論在火災(zāi)事故調(diào)查中的應(yīng)用初探[J];價(jià)值工程;2013年27期

7 韓智;;現(xiàn)場(chǎng)總體痕跡的研究及利用[J];公安論壇;1988年02期

8 嚴(yán)麗華;改進(jìn)《痕跡學(xué)》教學(xué)的思考[J];江蘇公安專(zhuān)科學(xué)校學(xué)報(bào);1994年02期

9 莫德升;論模擬痕跡教學(xué)法[J];公安教育;1997年06期

10 莫德升;論模擬痕跡教學(xué)法[J];政法學(xué)刊;1997年04期

相關(guān)會(huì)議論文 前1條

1 冉光平;;痕跡學(xué)在法醫(yī)學(xué)中的體會(huì)[A];全國(guó)第七次法醫(yī)學(xué)術(shù)交流會(huì)論文摘要集[C];2004年

相關(guān)碩士學(xué)位論文 前2條

1 楊敬樹(shù);剪切痕跡激光檢測(cè)信號(hào)自適應(yīng)匹配算法研究[D];昆明理工大學(xué);2017年

2 賈子庶;人體骨骼致傷工具推斷研究[D];山西醫(yī)科大學(xué);2010年

，

本文編號(hào)：2508907