国产bbaaaaa片,成年美女黄网站色视频免费,成年黄大片,а天堂中文最新一区二区三区,成人精品视频一区二区三区尤物

首頁> 外文學(xué)位 >Video-based animal behavior analysis.
【24h】

Video-based animal behavior analysis.

機(jī)譯：基于視頻的動(dòng)物行為分析。

獲取原文
獲取原文并翻譯 | 示例
  • 摘要
  • 著錄項(xiàng)
  • 相似文獻(xiàn)
  • 相關(guān)主題

摘要

It has become increasingly popular to study animal behaviors with the assistance of video recordings. The traditional way to do this is to first videotape the animal for a period of time, and then a human observer watches the video and annotates the behaviors of the animal manually. This is a time and labor consuming process. Moreover, the observation results vary among different observers. Thus it would be a great help if the behaviors could be accurately derived from an automated video processing and behavior analysis system. We are interested in developing techniques that will facilitate such a system for studying animal behaviors.;The video based behavior analysis systems can be decomposed into four major problems: behavior modeling, feature extraction from video sequences, basic behavior unit (BBU) discovery and complex behavior recognition. The recognition of basic and complex behaviors involves behavior definition, characterization and modeling. In the literature, there exist various techniques that partially address these problems for applications involving human motions and vehicle surveillance.;We propose a system approach to tackle these problems for animals. We first propose a behavior modeling framework, and a behavior model consisting of four levels: physical, physiological, contextual, and conceptual. Then we propose that the feature extraction and selection shall be guided by intrinsic variables that can distinguish different BBUs. BBUs are then determined from these features using the modified affinity graph method and a classification tree approach. We further investigated the application of a vector fusion method to reduce the feature dimensionality. Finally, we present results on analyzing behavior patterns for a simple problem, and apply the behavior models (transition probabilities, etc.) and rules (gained from prior knowledge) to correct and update the behaviors. These steps have been successfully applied to synthetic or real mouse video data, and in the future we expect to extend the methodology to study other video scenarios, like human behaviors or sports analysis.
機(jī)譯：在錄像的幫助下研究動(dòng)物行為已變得越來越普遍。傳統(tǒng)方法是先對(duì)動(dòng)物進(jìn)行錄像，然后再由人類觀察者觀看視頻并手動(dòng)注釋動(dòng)物的行為。這是一個(gè)耗時(shí)且費(fèi)力的過程。而且，觀察結(jié)果在不同觀察者之間也不同。因此，如果可以從自動(dòng)視頻處理和行為分析系統(tǒng)中準(zhǔn)確地得出行為，那將是一個(gè)很大的幫助。我們對(duì)開發(fā)有助于研究這種動(dòng)物行為的系統(tǒng)的技術(shù)感興趣?；谝曨l的行為分析系統(tǒng)可以分解為四個(gè)主要問題：行為建模，從視頻序列中提取特征，基本行為單位（BBU）發(fā)現(xiàn)和復(fù)雜行為識(shí)別?；拘袨楹蛷?fù)雜行為的識(shí)別涉及行為定義，表征和建模。在文獻(xiàn)中，存在多種技術(shù)可以部分解決這些問題，以解決涉及人體運(yùn)動(dòng)和車輛監(jiān)控的應(yīng)用。我們提出了一種系統(tǒng)方法來解決動(dòng)物的這些問題。我們首先提出一個(gè)行為建?？蚣埽约耙粋€(gè)包含四個(gè)層次的行為模型：物理，生理，上下文和概念。然后，我們建議特征提取和選擇應(yīng)以能夠區(qū)分不同BBU的固有變量為指導(dǎo)。然后使用改進(jìn)的親和圖方法和分類樹方法從這些特征中確定BBU。我們進(jìn)一步研究了矢量融合方法在減少特征維數(shù)方面的應(yīng)用。最后，我們提出分析一個(gè)簡(jiǎn)單問題的行為模式的結(jié)果，并應(yīng)用行為模型（轉(zhuǎn)換概率等）和規(guī)則（從先驗(yàn)知識(shí)中獲得）來糾正和更新行為。這些步驟已成功地應(yīng)用于合成或真實(shí)的鼠標(biāo)視頻數(shù)據(jù)，并且在將來，我們希望擴(kuò)展該方法以研究其他視頻場(chǎng)景，例如人類行為或運(yùn)動(dòng)分析。

著錄項(xiàng)

相似文獻(xiàn)

  • 外文文獻(xiàn)
  • 中文文獻(xiàn)
  • 專利
獲取原文

客服郵箱：kefu@zhangqiaokeyan.com

京公網(wǎng)安備：11010802029741號(hào) ICP備案號(hào)：京ICP備15016152號(hào)-6 六維聯(lián)合信息科技 (北京) 有限公司?版權(quán)所有

  • 客服微信

  • 服務(wù)號(hào)