Description: 摘要:介绍了一个数码管显示数字仪表表盘读数的自动识别系统，在字符提取阶段，采用颜色提取的方法，使得尽管图片的字体颜
色与背景颜色相同，但只要亮度不同，也能很好的提取出字符部分。在字符分割过程中，采用增益方法识别边界，准确分割出单个字符。
并采用统计法和扫描法相结合的方法来识别数字.实验结果表明提出的方案可行，具有较强的鲁棒性.-Abstract: This paper presents a digital display digital meter dial readings of the automatic identification system, in the character extraction stage, using color extraction method, although the picture makes the font color and background color, but as long as the brightness of different, but also a very good extraction part of the characters. In the character segmentation process, the method used to identify the boundary gain, accurate segmentation of individual characters. And using statistical method and scanning method combined to identify the figures. The experimental results show that our proposal is feasible and has strong robustness. Platform: |
Size: 304128 |
Author:casey1986 |
Hits:
Description: 可以将分割完的子图像进行二值化，二值化效果很明显，易于下一步的分割识别-You can split end of the sub-image binarization, thresholding effect is obvious and easy to identify the next stage of segmentation Platform: |
Size: 2048 |
Author:梁勇 |
Hits:
Description: This paperpresent a novel framework for inferring global behaviour patterns through modelling behaviour correlations in a wide-area scene and detecting any
anomaly in behaviours occurring both locally and globally. Specifically,
This paperpropose a semantic scene segmentation model to decompose a wide-area
scene into regions where behaviours share similar characteristic and are represented as classes of video events bearing similar features. To model behavioural correlations globally, This paperinvestigate both a probabilistic Latent Semantic Analysis (pLSA) model and a two-stage hierarchical pLSA model for
global behaviour inference and anomaly detection. The proposed framework
is validated by experiments using complex crowded outdoor scenes. Platform: |
Size: 497664 |
Author:shpark |
Hits:
Description: In this thesis we present an operational computer vision system for real-time detection and tracking of human motion. The system captures monocular video of a scene and identifies those moving objects which are characteristically human. This serves as both a proof-ofconcept and a verification of other existing algorithms for human motion detection. An approach to statistical modeling of motion developed by Y. Song is coupled with a preprocessing stage of image segmentation and point feature tracking. This design allows a system that is robust with respect to occlusion, clutter, and extraneous motion. The results of experiments with the system indicate the ability to minimize both false detections and missed detections. Platform: |
Size: 778240 |
Author:linuszhao |
Hits:
Description: 统一资源定位符（URL）是网站页面的地址判别方式，也是蜘蛛抓取网站网页信息的途径。那搜索引擎蜘蛛是如何通过URL链接抓取网站页面的呢？搜索引擎工作大致分为三个阶段：爬行和抓取阶段（搜索引擎蜘蛛访问页面，并获取页面html代码存入数据库）：预处理（对页面文字进行提取、分词、消除噪音、去重 和建立索引）；排名（根据页面的相关性和网站权重高低展示给用户）。-Uniform Resource Locator (URL) address discrimination is the way web pages, but also the way spiders crawl the website pages of information. That is how the search engine spiders to crawl web pages by URL link it? Search engines work is broadly divided into three stages: crawling and crawling stage (search engine spiders to access the page, and the page html code to get into the database): pretreatment (on page text extraction, segmentation, eliminate noise, and to re-establish index) ranking (based on the correlation of the page and site level weights presented to the user). Platform: |
Size: 4096 |
Author:smith |
Hits:
Description: 本站采用的是一种用最小切分和逆向最大匹配法结合的一种复合分词算法，目前对岐义识别还处于初步研究阶段.-Powered by is a composite word segmentation algorithm with a minimum and maximum matching method combined reverse, the current recognition of ambiguity still in the preliminary research stage. Platform: |
Size: 676864 |
Author:vhnd |
Hits:
Search - stage by stage segmentation