發(fā)布時(shí)間：2018-06-15 10:55

  本文選題：視覺(jué)監(jiān)控 + 布爾核函數(shù)��； 參考：《電氣應(yīng)用》2014年22期

【摘要】：在長(zhǎng)距離的電力運(yùn)輸中,對(duì)于漏電的監(jiān)測(cè)效率較低。提出基于布爾核SVM識(shí)別算法的大型智能視覺(jué)監(jiān)控下漏電監(jiān)測(cè)方法。通過(guò)監(jiān)控設(shè)備拍攝漏電發(fā)生時(shí)的電火花圖像,對(duì)其圖像進(jìn)行顏色和形狀參數(shù)特征的提取,構(gòu)建布爾核SVM模型,完成監(jiān)控設(shè)備拍攝下的漏電發(fā)生過(guò)程中電火花圖像的基本識(shí)別。實(shí)驗(yàn)結(jié)果表明,利用該算法進(jìn)行大型智能視覺(jué)監(jiān)控下的漏電監(jiān)測(cè)軟件設(shè)計(jì),能夠極大地提高識(shí)別能力和識(shí)別的準(zhǔn)確率,及時(shí)監(jiān)測(cè)到漏電情況,保證了用電區(qū)域的安全。
[Abstract]:In long distance electric transportation, the efficiency of monitoring leakage is low. This paper presents a large scale intelligent visual monitoring method for leakage monitoring based on Boolean kernel SVM recognition algorithm. By using the monitoring equipment to capture the EDM image and extract the color and shape parameters of the EDM image, a Boolean kernel SVM model is constructed to recognize the EDM image in the process of the EDM generated by the monitoring equipment. The experimental results show that using this algorithm to design the leakage monitoring software under large-scale intelligent vision monitoring can greatly improve the recognition ability and the accuracy of recognition, monitor the leakage situation in time, and ensure the safety of the electric power area.
【作者單位】： 寧夏大學(xué)物理電氣信息學(xué)院;
【分類號(hào)】：TM934.31;TP391.41
，

本文編號(hào)：2021773