PSYCH OpenIR  > 社会与工程心理学研究室
图像侦查中的自然人机交互工具及其评价
Alternative TitleNatural Interaction Tool in Video Investigation and Its Evaluation
郑妍1,2; 张玉麟1,2; 赵国朕1,2
2018-10
Conference Name第二十一届全国心理学学术会议
Correspondent Emailzhaogz@psych.ac.cn
Source Publication摘要集-第二十一届全国心理学学术会议
Pages809-810
Conference Date10.30-11.2
Conference Place北京
Contribution Rank1
Abstract

摘要:像侦查技术是通过侦查视频图像锁定犯罪嫌疑人的一种侦查技术手段,通过调用可能包含嫌疑人的视频监控并逐个查阅筛选明显的线索,以此确定嫌疑人的行踪。现有的图侦技术仍然依赖于大量的人力投入,警员往往要长时间的连续观看海量的视频图像,并且在整个查阅视频的过程中,警员只能通过纸笔记录案件要素和线索,不利于要素的梳理和串联分析。首先,本研究基于对一线图像侦查人员的访谈,绘制了警员对于图像侦查工具的心理模型,并开发与图像侦查心理模型相符的视频交互技术,支持草图标注,线索超链接,自动轨迹绘制等功能,以简化警员的工作流程,提高工作的效率。同时,为了验证该图像侦查交互技术的可用性,本研究选取了30名学生被试(15男15女,平均年龄24岁)并随机分为两组(实验组和对照组),来完成与图像侦查任务相关的三个子任务,同时记录行为绩效指标和反映心理负荷的脑电数据。其中,实验组使用的是根据心理模型设计的交互工具,对照组使用的是原工具,即普通播放器配合纸笔记录的工作方式。任务一要求被试观看视频并搜索目标,使用的视频材料是由公安局提供的真实的案件监控视频片段。结果表明实验组完成任务用时显著短于对照组,并且在F4和F5处和额叶平均的theta波功率显著小于控制组。任务二要求被试将视频中看到的目标行驶方向画在地图上,任务三是线索回溯任务,结果均表明实验组用时显著短于对照组,但脑电指标差异不显著。总的来说,根据用户心理模型设计的图像侦查交互工具可以提高任务绩效,降低心理负荷。

Other Abstract

Abstract: Video detection is a kind of investigation technique to detect criminal suspects by detecting video images. By watching the surveillance videos that may contain suspects, selecting the relevant clues from videos, video investigation analysts can identify suspects’ trajectory. However, existing video investigation work still relies on a large number of human labor. Analysts often watch massive surveillance videos for a long time. And in the whole process of video search, the police can only record clues through paper and pen. Poor relevance of video material and text clue hinders video investigation analysis. Based on the interview with video investigation analysts, their mental model of the video detection tool was drew. Based on the mental model, a video investigation tool was realized with functions of sketch annotation, hyperlink and trajectory drawing. In order to test the usability of this video investigation tool, 30 participants were recruited (15 male and 15 female, mean age was 24) and were randomly divided into two groups (an experimental group and a control group) to complete three sub tasks which were related to the image detection task. Behavioral performance indicators and EEG indicators which reflect mental workload were recorded at the same time. The experimental group used the video investigation tool and the control group used the original tool. Task1 requires the subjects to watch videos and search for targets. The video material used is a real case monitoring video clip provided by the Public Security Bureau. Results showed that the task completion time of the experimental group was significantly shorter than the control group. Compared to the control group, significant lower theta power was found at F4, F5 as well as the average power of the frontal lobe in the experimental group. Task2 asked subjects to draw the direction of the target in the video on the map and task3 was a trail backtracking task. Results showed that the task completion time of the experimental group was significantly shorter than the control group. However, no significant difference was found in EEG index. In general, the video investigation tool designed based on user mental models can improve task performance and reduce mental workload.

Keyword图像侦查 交互自然性 心理模型 心理负荷
Language中文
Document Type会议论文
Identifierhttp://ir.psych.ac.cn/handle/311026/27300
Collection社会与工程心理学研究室
Affiliation1.中国科学院心理研究所
2.中国科学院大学心理系
First Author AffilicationInstitute of Psychology, Chinese Academy of Sciences
Recommended Citation
GB/T 7714
郑妍,张玉麟,赵国朕. 图像侦查中的自然人机交互工具及其评价[C],2018:809-810.
Files in This Item:
There are no files associated with this item.
Related Services
Recommend this item
Bookmark
Usage statistics
Export to Endnote
Google Scholar
Similar articles in Google Scholar
[郑妍]'s Articles
[张玉麟]'s Articles
[赵国朕]'s Articles
Baidu academic
Similar articles in Baidu academic
[郑妍]'s Articles
[张玉麟]'s Articles
[赵国朕]'s Articles
Bing Scholar
Similar articles in Bing Scholar
[郑妍]'s Articles
[张玉麟]'s Articles
[赵国朕]'s Articles
Terms of Use
No data!
Social Bookmark/Share
All comments (0)
No comment.
 

Items in the repository are protected by copyright, with all rights reserved, unless otherwise indicated.