本文選題：林火識別 + 圖像處理�。� 參考：《北京林業(yè)大學(xué)》2016年碩士論文

【摘要】：森林火災(zāi)的發(fā)生對經(jīng)濟(jì)、生態(tài)環(huán)境等會產(chǎn)生巨大影響。世界各國對林火監(jiān)測系統(tǒng)的研究都非常重視。傳感器監(jiān)測方法受環(huán)境影響較大,不適宜應(yīng)用在大尺度空間的森林火災(zāi)監(jiān)測上。傳統(tǒng)圖像型監(jiān)測方法需要對圖像進(jìn)行必要處理,人工提取特征,特征選擇成為了能否達(dá)到理想效果的關(guān)鍵因素。針對上述情況,本文將林火識別與機(jī)器學(xué)習(xí)領(lǐng)域的深度學(xué)習(xí)算法相結(jié)合,將其中的卷積神經(jīng)網(wǎng)絡(luò)模型應(yīng)用在林火識別上。深度網(wǎng)絡(luò)可以自動提取輸入圖像特征,通過層與層之間的傳遞,將底層特征組合形成高層的抽象特征,避免了傳統(tǒng)方法中人工提取特征的復(fù)雜性和盲目性。深度卷積神經(jīng)網(wǎng)絡(luò)特有的局部感受野和權(quán)值共享技術(shù)減少了參數(shù)的數(shù)目,降低了算法訓(xùn)練的難度,下采樣的使用增強(qiáng)了網(wǎng)絡(luò)容忍圖像畸變的能力。實驗結(jié)果證明,該方法取得了較為理想的效果。本文的主要工作如下：(1)通過實驗和網(wǎng)上搜集構(gòu)建林火數(shù)據(jù)庫。(2)人工提取圖像特征,使用目前比較常用的方法如支持向量機(jī)、徑向基函數(shù)網(wǎng)絡(luò)、BP神經(jīng)網(wǎng)絡(luò)對特征進(jìn)行識別,并對實驗結(jié)果進(jìn)行分析。(3)在對卷積神經(jīng)網(wǎng)絡(luò)深入研究的基礎(chǔ)上,對森林火災(zāi)圖像進(jìn)行有無火災(zāi)的辨別,針對夜晚和白天背景不同的情況,設(shè)計不同的網(wǎng)絡(luò)進(jìn)行識別,對不同的結(jié)構(gòu),不同的參數(shù)進(jìn)行比較分析。最終得到的夜晚林火識別模型正確率為95.71%,白天林火識別模型正確率為98%,與人工提取特征的傳統(tǒng)圖像型監(jiān)測方法相比,具有顯著優(yōu)勢。
[Abstract]:The occurrence of forest fires has a great impact on the economy, the ecological environment and so on. All countries of the world have paid great attention to the research of forest fire monitoring system. The sensor monitoring method is greatly influenced by the environment and is not suitable for application in the monitoring of forest fire in large scale space. The traditional image monitoring method needs the necessary processing of the image and artificial extraction. Feature selection is the key factor to achieve the ideal effect. In this case, this paper combines the forest fire recognition with the depth learning algorithm in the machine learning field, and applies the convolution neural network model in the forest fire recognition. The depth network can automatically extract the feature of the input image, through the layer and the layer. In order to avoid the complexity and blindness of the artificial extraction of the traditional methods, the specific local receptive field and weight sharing technology of the deep convolution neural network reduce the number of parameters and reduce the difficulty of the algorithm training. The use of lower sampling enhances the network tolerance of image distortion. The experimental results show that the method has achieved better results. The main work of this paper is as follows: (1) build forest fire database through experiment and online collection. (2) artificial extraction of image features, using the commonly used methods such as support vector machine, radial basis function network, BP neural network to identify the characteristics, and the actual The results are analyzed. (3) on the basis of the thorough research on the convolution neural network, there is no fire discrimination on the forest fire images. According to the different circumstances in the night and the daytime, different networks are designed and the different structures and different parameters are compared and analyzed. The final recognition model of the night forest fire is correct. The accuracy rate of forest fire recognition model is 98% during the daytime, and has a significant advantage compared with the traditional image monitoring method based on artificial feature extraction. 95.71%.
【學(xué)位授予單位】：北京林業(yè)大學(xué)
【學(xué)位級別】：碩士
【學(xué)位授予年份】：2016
【分類號】：TP391.41

【參考文獻(xiàn)】

相關(guān)期刊論文 前10條

1 李睿凡;陳佳潔;周延泉;王小捷;鐘義信;;深度學(xué)習(xí)中卷積神經(jīng)網(wǎng)絡(luò)的教學(xué)探討[J];計算機(jī)教育;2015年18期

2 范一舟;馬洪兵;;基于視頻的林火煙霧識別方法[J];清華大學(xué)學(xué)報(自然科學(xué)版);2015年02期

3 郭偉;昂海松;張思玉;;小型無人機(jī)林火監(jiān)測與圖像處理技術(shù)[J];森林防火;2014年04期

4 何清;李寧;羅文娟;史忠植;;大數(shù)據(jù)下的機(jī)器學(xué)習(xí)算法綜述[J];模式識別與人工智能;2014年04期

5 魯向擁;陳悅;張瞳;葛貝;;Deep Learning算法分析和模型改進(jìn)[J];計算機(jī)光盤軟件與應(yīng)用;2014年06期

6 鄭胤;陳權(quán)崎;章毓晉;;深度學(xué)習(xí)及其在目標(biāo)和行為識別中的新進(jìn)展[J];中國圖象圖形學(xué)報;2014年02期

7 Fei-wei QIN;Lu-ye LI;Shu-ming GAO;Xiao-ling YANG;Xiang CHEN;;A deep learning approach to the classification of 3D CAD models[J];Journal of Zhejiang University-Science C(Computers & Electronics);2014年02期

8 余凱;賈磊;陳雨強(qiáng);徐偉;;深度學(xué)習(xí)的昨天、今天和明天[J];計算機(jī)研究與發(fā)展;2013年09期

9 楊昭;高雋;謝昭;吳克偉;;局部Gist特征匹配核的場景分類[J];中國圖象圖形學(xué)報;2013年03期

10 李海峰;李純果;;深度學(xué)習(xí)結(jié)構(gòu)和算法比較分析[J];河北大學(xué)學(xué)報(自然科學(xué)版);2012年05期

相關(guān)碩士學(xué)位論文 前1條

1 郭歡;基于數(shù)字圖像處理的森林火災(zāi)探測算法的研究[D];東北林業(yè)大學(xué);2013年

，

本文編號：2023598