基于音頻特征和模糊神經(jīng)網(wǎng)絡(luò)的禽流感病雞檢測(cè)

張鐵民，黃俊端

?

基于音頻特征和模糊神經(jīng)網(wǎng)絡(luò)的禽流感病雞檢測(cè)

張鐵民1,2，黃俊端1

（1. 華南農(nóng)業(yè)大學(xué)工程學(xué)院，廣州 510642； 2. 華南農(nóng)業(yè)大學(xué)國(guó)家生豬種業(yè)工程技術(shù)研究中心，廣州 510642）

為了能在早期發(fā)現(xiàn)禽流感并進(jìn)行預(yù)防，該文提出了一種基于音頻特征和模糊神經(jīng)網(wǎng)絡(luò)的禽流感病雞檢測(cè)方法。依據(jù)獲取的家禽音頻和環(huán)境及其他噪聲的譜熵差別大的特點(diǎn)，在復(fù)雜環(huán)境中分析并提取出雞聲，丟棄非雞聲段，對(duì)提取的雞聲進(jìn)行分析及處理，計(jì)算短時(shí)過零率、短時(shí)能量以及短時(shí)過零率與短時(shí)能量混合特征，用作判別患禽流感的病雞和健康雞的依據(jù)。利用T-S模糊神經(jīng)網(wǎng)絡(luò)，對(duì)提取出來的家禽音頻特征進(jìn)行訓(xùn)練和識(shí)別，試驗(yàn)表明隸屬度函數(shù)為鐘形函數(shù)、隸屬度個(gè)數(shù)為2時(shí)模糊神經(jīng)網(wǎng)絡(luò)對(duì)試驗(yàn)提取的3個(gè)雞聲特征組成的3組測(cè)試集的敏感性分別為75.47%、80.39%和76.92%，特異性分別為80.85%、79.59%和72.92%，正確識(shí)別率分別為78%、80%和75%。該研究為規(guī)?；仪蒺B(yǎng)殖場(chǎng)及大型家禽流通市場(chǎng)的禽流感病禽識(shí)別提供一套快速、高效檢測(cè)方法。

神經(jīng)網(wǎng)絡(luò)；識(shí)別；提取；譜熵；短時(shí)過零率；短時(shí)能量；雞病檢測(cè)

0 引 言

隨著現(xiàn)代規(guī)?；B(yǎng)雞業(yè)的發(fā)展，禽流感對(duì)經(jīng)濟(jì)、食品安全和人類健康有著重要的影響[1]，因此在規(guī)?；B(yǎng)殖環(huán)境中及時(shí)對(duì)患有禽流感的病雞進(jìn)行快速、準(zhǔn)確的識(shí)別不僅直接關(guān)系到養(yǎng)雞業(yè)的經(jīng)濟(jì)效益，同時(shí)對(duì)預(yù)防禽流感交叉?zhèn)鞑ゾ哂兄匾饬x。傳統(tǒng)的雞病診斷主要依靠獸醫(yī)對(duì)雞的姿態(tài)、雞冠、聲音以及糞便進(jìn)行觀察[2-4]，費(fèi)時(shí)費(fèi)力，尤其對(duì)于規(guī)?；B(yǎng)殖場(chǎng)，效率低下。然而隨著現(xiàn)代電子計(jì)算機(jī)技術(shù)的發(fā)展，許多新的、更高效的雞病檢測(cè)方法被提出來。

目前多采用視頻角度,許多基于機(jī)器視覺對(duì)其他動(dòng)物行為監(jiān)測(cè)的方法被陸續(xù)提出[5-8]。而對(duì)于運(yùn)用計(jì)算機(jī)視覺識(shí)別病雞、監(jiān)測(cè)雞的行為，畢敏娜等[9-11]通過對(duì)家禽姿態(tài)識(shí)別進(jìn)行病雞識(shí)別，其支持向量機(jī)模型在測(cè)試集的識(shí)別率達(dá)到99.469%。王琳等[12]用數(shù)值積分方法提取出雞的深度圖像特征，結(jié)合神經(jīng)網(wǎng)絡(luò)，實(shí)現(xiàn)群體肉雞的質(zhì)量估計(jì)。勞風(fēng)丹等[13]用機(jī)器視覺實(shí)現(xiàn)對(duì)單只蛋雞的行為識(shí)別，監(jiān)測(cè)其生產(chǎn)和健康狀況。而從音頻角度看，動(dòng)物的發(fā)聲包含豐富的信息，能夠在一定程度上反饋動(dòng)物的健康情況[14-16]，目前有許多基于音頻分析的方法應(yīng)用于其他動(dòng)物的行為和健康監(jiān)測(cè)[17-22]。而對(duì)于分析雞的音頻研究，Banakar Ahmad等[23]用數(shù)據(jù)挖掘方法和Dempster-Shafer證據(jù)理論，結(jié)合支持向量機(jī)作為識(shí)別工具，識(shí)別和分類幾種常見的雞病。國(guó)內(nèi)曹晏飛[24]針對(duì)棲架飼養(yǎng)模式下蛋雞發(fā)出的聲音，提出了基于功率譜密度特征的分類識(shí)別方法，該方法的平均識(shí)別率達(dá)到95%。余禮根等[25]以海蘭褐蛋雞為例，收集了其在小規(guī)模（5只）飼養(yǎng)條件下的叫聲信息，提取其包括持續(xù)時(shí)間、基音頻率、頻譜質(zhì)心、共振峰及其衍生的發(fā)聲特征參數(shù)，構(gòu)建出蛋雞發(fā)聲音頻數(shù)據(jù)庫(kù)，分析蛋雞發(fā)聲和其他行為的聯(lián)系，得出分析蛋雞發(fā)聲特征有助于了解其行為特性、機(jī)體狀態(tài)以及種群間的信息傳遞。

本文立足于應(yīng)用音頻分析技術(shù)，通過分析籠養(yǎng)雞中的健康和感染禽流感病毒的雞叫聲，從音頻角度，提出一種在環(huán)境噪聲背景中提取出雞聲的有效音頻特征的方法，并用模糊神經(jīng)網(wǎng)絡(luò)作為分類器，識(shí)別禽流感病雞的叫聲和健康雞的叫聲，以期為家禽養(yǎng)殖業(yè)提供一種非接觸式的自動(dòng)化檢測(cè)禽流感的方法。

1 健康和患禽流感病的雞聲獲取方法

雞聲獲取方法：第一天：12:00將試驗(yàn)用的14只五周齡的無特定病原（specific pathogen-free，SPF）白來航雞放入雞隔離器中，讓其熟悉生存環(huán)境，減少應(yīng)激反應(yīng)；第二天，16:30將錄音筆放入置于雞隔離器中，記錄雞聲，如圖1所示，錄音筆為T&F-91加強(qiáng)版32G數(shù)字高清錄音筆，采樣頻率48 000 Hz。錄音筆能長(zhǎng)時(shí)間連續(xù)不間斷錄音120 h；第三天16:00對(duì)雞進(jìn)行感染，感染后的雞呈現(xiàn)出眼瞼水腫，精神呆滯，聲音嘶啞，羽毛蓬松等特征。錄音筆持續(xù)記錄雞的叫聲直到第六天感染禽流感病毒雞全部死亡。取感染禽流感病毒前一天的雞的叫聲作為健康雞的叫聲（試驗(yàn)的第二天），取感染禽流感病毒后的雞的叫聲作為禽流感病雞的叫聲（試驗(yàn)的第四天）。

感染方法：采用的H7N9亞型禽流感病毒由華南農(nóng)業(yè)大學(xué)獸醫(yī)學(xué)院禽病研究室分離鑒定，在華南農(nóng)業(yè)大學(xué)動(dòng)物生物安全三級(jí)（animal biosafety level 3，ABSL-3）實(shí)驗(yàn)室中進(jìn)行，所有操作均按照ABSL-3相關(guān)標(biāo)準(zhǔn)步驟及相關(guān)生物安全標(biāo)準(zhǔn)進(jìn)行。利用Reed-Muench法計(jì)算病毒半數(shù)胚胎感染劑量（50% Embryo Infective Dose，EID50），用含10 000/mL青霉素和鏈霉素的無菌PBS將H7N9亞型禽流感病毒均稀釋到106EID50/0.1 mL，每只雞經(jīng)滴鼻點(diǎn)眼感染106EID50/0.1 mL病毒稀釋液0.1 mL。

圖1 音頻采集環(huán)境

2 雞聲信號(hào)的時(shí)域分析與處理

2.1 預(yù)加重

由于聲音信號(hào)低頻信噪比將大，而高頻信噪比不足，需對(duì)輸入的數(shù)字雞聲信號(hào)的高頻部分進(jìn)行預(yù)加重處理，采用高通濾波器進(jìn)行預(yù)加重，以提高雞聲的高頻分辨率，高通濾波器的傳遞函數(shù)如下[26-27]。

其中為預(yù)加重系數(shù)，0.9<<1.0。設(shè)時(shí)刻語音采樣值為()，經(jīng)過預(yù)加重處理后的結(jié)果為()=()?(?1)，這里取0.98。

圖2為原始含噪雞聲信號(hào)和預(yù)加重后的頻率幅值曲線，經(jīng)過高通濾波器后，雞聲的低頻部分被削弱，高頻部分被增強(qiáng)。原始含噪雞聲信號(hào)和預(yù)加重后的含噪雞聲信號(hào)都經(jīng)過歸一化處理。

圖2 預(yù)加重效果

2.2 雞聲信號(hào)的加窗與分幀

雞聲信號(hào)是時(shí)變信號(hào)，但在一個(gè)短時(shí)間內(nèi)（10～30 ms），其特性基本保持不變，可以將其看做準(zhǔn)穩(wěn)態(tài)過程，其信號(hào)具有短時(shí)平穩(wěn)性[27-30]。為了使截取的雞聲信號(hào)波形緩慢降為零，減小雞聲幀的截?cái)嘈?yīng)[24]，選hamming窗對(duì)雞聲信號(hào)分幀，取21.3 ms為一幀，hamming窗數(shù)學(xué)表達(dá)式如下[26-27]，表示幀長(zhǎng)量。

2.3 基于譜熵法的雞聲端點(diǎn)檢測(cè)

熵表示信息的有序程度，由Shannon引用到信息理論中來，信號(hào)以信息熵來作為信息選擇和不確定性的度量[31]。Shen等[32]在試驗(yàn)中發(fā)現(xiàn)語音的熵和噪聲的熵存在較大的差異，首次提出基于熵的語音端點(diǎn)檢測(cè)方法。雞叫聲的熵跟環(huán)境噪聲的熵有明顯不同，提出一種基于譜熵法的雞聲端點(diǎn)檢測(cè)，從一段含噪雞聲中準(zhǔn)確地找出雞聲信號(hào)起始點(diǎn)和結(jié)束點(diǎn)，使有效的雞聲信號(hào)和無用的噪聲信號(hào)得以分離。

2.3.1 基于譜熵法的雞聲端點(diǎn)檢測(cè)算法

1）對(duì)含噪雞聲進(jìn)行加窗分幀處理，計(jì)算每一幀的譜的能量。含噪雞聲信號(hào)定義為()，加hamming分幀處理后得到的第幀含噪雞聲信號(hào)為x()，其快速傅里葉變換(Fast Fourier transform，F(xiàn)FT)表示為X()，其轉(zhuǎn)置矩陣表示為X()，下標(biāo)表示為第幀，表示第條譜線，表示FFT的點(diǎn)數(shù)，取=1 024。每一幀聲音信號(hào)在頻域中的短時(shí)能量E為

2）計(jì)算每一幀中每個(gè)樣本點(diǎn)的概率密度函數(shù)。定義某一譜線的能量譜為Y()=X()X()，則每個(gè)頻率分量的歸一化譜概率密度p()為

3）計(jì)算每一幀的譜熵值H。

4）設(shè)定判決門限進(jìn)行端點(diǎn)檢測(cè)。本文選擇判決門限為含噪雞聲所有幀的譜熵值的平均值減去所有幀中譜熵值的最小值。

2.3.2 基于譜熵法的雞聲端點(diǎn)檢測(cè)效果

基于譜熵法的雞聲端點(diǎn)檢測(cè)流程如圖3所示。

圖3 基于譜熵法的雞聲端點(diǎn)檢測(cè)流程

本文選擇最小雞聲長(zhǎng)度為3幀?；谧韵嚓P(guān)函數(shù)的雞聲端點(diǎn)檢測(cè)效果如圖4所示，圖中紅色實(shí)線表示雞聲起始幀位置，藍(lán)色虛線表示雞聲結(jié)束幀位置。

注：紅色實(shí)線表示端點(diǎn)檢測(cè)中雞聲起始幀的位置，藍(lán)色虛線表示端點(diǎn)檢測(cè)中雞聲結(jié)束幀的位置。

3 雞聲短時(shí)過零率和短時(shí)能量

3.1 雞聲短時(shí)過零率計(jì)算

短時(shí)過零率（Short-time zero-crossing rate，STZ）表示聲音信號(hào)波形穿過橫軸的次數(shù)，對(duì)于離散信號(hào)如果相鄰的取樣值改變符號(hào)則稱為過零，一幀雞聲信號(hào)x()的短時(shí)過零率Z的計(jì)算為

式中下標(biāo)表示第幀雞聲信號(hào)。sgn[ ]是求符號(hào)函數(shù)，即

為了消除錄音器隨機(jī)微弱電流噪聲的影響，引入一個(gè)去噪變量，本文選擇去噪變量為0.000 1。則雞聲信號(hào)的短時(shí)過零率Z的計(jì)算為

3.2　雞聲短時(shí)能量計(jì)算

短時(shí)能量（short-time energy，STE）用來度量音頻信號(hào)的幅度值變化，雞在感染禽流感后叫聲發(fā)生改變，聲音能量也發(fā)生改變，一幀雞聲x()短時(shí)能量E的計(jì)算為

3.3 雞聲短時(shí)過零率與短時(shí)能量混合特征

雞聲短時(shí)過零率包含雞聲信號(hào)的符號(hào)信息，雞聲短時(shí)能量包含雞聲信號(hào)的幅度信息，將雞聲短時(shí)過零率與雞聲短時(shí)能量的數(shù)值相乘，作為同時(shí)包含雞聲信號(hào)的符號(hào)信息和雞聲信號(hào)的幅度信息的雞聲短時(shí)過零率與短時(shí)能量混合特征，一幀雞聲信號(hào)x()的短時(shí)過零率與短時(shí)能量混合特征K的計(jì)算為

4 基于模糊神經(jīng)網(wǎng)絡(luò)的雞聲識(shí)別

神經(jīng)網(wǎng)絡(luò)具有并行計(jì)算，分布式信息存儲(chǔ)，容錯(cuò)能量強(qiáng)及自適應(yīng)學(xué)習(xí)能力等優(yōu)勢(shì)，模糊邏輯是一種處理不確定性和非線性的強(qiáng)有力的工具，模糊神經(jīng)網(wǎng)絡(luò)將神經(jīng)網(wǎng)絡(luò)與模糊邏輯結(jié)合起來，具備兩者的長(zhǎng)處，性能比單純的神經(jīng)網(wǎng)絡(luò)或者單純的模糊邏輯更強(qiáng)[33]。

4.1 T-S模糊模型

模糊模型主要有2種，一種是模糊規(guī)則的后件是輸出量的某一模糊集合，稱為模糊系統(tǒng)的標(biāo)準(zhǔn)模型，另一種是模糊規(guī)則的后件輸入是輸入語言變量函數(shù)，由Takagi等提出[34-36]，稱為T-S模糊模型。模糊系統(tǒng)的標(biāo)準(zhǔn)模型雖然符合人的思維和語言表達(dá)習(xí)慣，但存在計(jì)算復(fù)雜、不利于數(shù)學(xué)分析等缺點(diǎn)。本文選用T-S模糊模型。

T-S模糊模型用如下的if-then規(guī)則形式定義，在規(guī)則R的情況下，模糊推理為

根據(jù)模糊計(jì)算結(jié)果計(jì)算模糊模型的歸一化輸出值y

4.2 T-S模糊神經(jīng)網(wǎng)絡(luò)

5 結(jié)果與分析

以雞聲短時(shí)過零率、短時(shí)能量和短時(shí)過零率與短時(shí)能量混合特征作為識(shí)別特征，取健康雞的識(shí)別特征和禽流感病雞的識(shí)別特征各450個(gè)，組成一個(gè)行數(shù)為900，列數(shù)為3的識(shí)別特征矩陣。構(gòu)造行數(shù)為900，列數(shù)為1的矩陣作為標(biāo)志矩陣，在標(biāo)志矩陣中，禽流感病雞聲用1表示，健康雞聲用0表示，將和組合在一起作為數(shù)據(jù)集（，），訓(xùn)練樣本為（x，y），x表示第個(gè)雞聲的識(shí)別特征，y表示第個(gè)雞聲是否患病的標(biāo)志（=1,2，…900），隨機(jī)打亂樣本的順序，取前600個(gè)樣本作為訓(xùn)練集，后300個(gè)樣本作為3組測(cè)試集，每組測(cè)試集100個(gè)樣本，各數(shù)據(jù)集中禽流感病雞聲特征和健康雞聲特征數(shù)量如表1所示。

表1 訓(xùn)練集和測(cè)試集的禽流感病雞聲特征和健康雞聲特征

圖5 模糊神經(jīng)網(wǎng)絡(luò)對(duì)測(cè)試集的識(shí)別率與訓(xùn)練次數(shù)的關(guān)系曲線

由圖5可知，模糊神經(jīng)網(wǎng)絡(luò)在訓(xùn)練次數(shù)為16時(shí)對(duì)測(cè)試集的識(shí)別率達(dá)到穩(wěn)定，隸屬度函數(shù)為鐘形函數(shù)的識(shí)別率最高。選擇隸屬度函數(shù)為鐘形函數(shù)，訓(xùn)練次數(shù)為16次，分別計(jì)算該模糊神經(jīng)網(wǎng)絡(luò)對(duì)3組測(cè)試集識(shí)別結(jié)果的8個(gè)統(tǒng)計(jì)值：正確的正例、正確的反例、錯(cuò)誤的正例、錯(cuò)誤的反例、敏感性、特異性、正確識(shí)別率和錯(cuò)誤識(shí)別率。本文中，敏感性為某一測(cè)試集中被正確診斷為禽流感病雞聲的個(gè)數(shù)與該測(cè)試集中所有禽流感病雞聲的數(shù)量之比，特異性為某一測(cè)試集被正確診斷為健康雞聲的個(gè)數(shù)與該測(cè)試集中所有健康雞聲的數(shù)量之比，敏感性和特異性的計(jì)算公式如式（18），（19）所示。數(shù)據(jù)統(tǒng)計(jì)結(jié)果如表2所示。

由表2可知，隸屬度函數(shù)為鐘形函數(shù)，隸屬度個(gè)數(shù)為2時(shí)，模糊神經(jīng)網(wǎng)絡(luò)對(duì)本試驗(yàn)提取的3個(gè)雞聲特征組成的3組測(cè)試集的敏感性分別為75.47%、80.39%和76.92%，特異性分別為80.85%、79.59%和72.92%，正確識(shí)別率分別為78%、80%和75%。由表2可知模糊神經(jīng)網(wǎng)絡(luò)對(duì)測(cè)試集的識(shí)別率最高達(dá)80%，識(shí)別率在75%到80%之間。

6 結(jié) 論

本文通過在ABSL-3實(shí)驗(yàn)室中對(duì)5周齡的SPF雞做禽流感病毒感染試驗(yàn)，收集了健康雞的叫聲和禽流感病雞的叫聲。以收集到的聲音數(shù)據(jù)為分析對(duì)象，首先對(duì)聲音信息進(jìn)行預(yù)處理、加窗和分幀，接著分析含噪雞聲的譜熵特征，提出了一種基于譜熵法的雞聲端點(diǎn)檢測(cè)方法，在含有噪聲的雞聲錄音中截取出雞的叫聲，舍棄環(huán)境噪聲。

通過計(jì)算有效雞聲幀的短時(shí)過零率、短時(shí)能量以及短時(shí)過零率與短時(shí)能量混合特征，使用基于T-S模糊模型的模糊神經(jīng)網(wǎng)絡(luò)做健康雞和禽流感病雞的雞聲特征識(shí)別，試驗(yàn)表明隸屬度函數(shù)為鐘形函數(shù)、隸屬度個(gè)數(shù)為2時(shí)模糊神經(jīng)網(wǎng)絡(luò)對(duì)本試驗(yàn)提取的3個(gè)雞聲特征組成的3組測(cè)試集的敏感性分別為75.47%、80.39%和76.92%，特異性分別為80.85%、79.59%和72.92%，正確識(shí)別率分別為78%、80%和75%。

本文所提出的雞聲特征提取和識(shí)別方法對(duì)家禽養(yǎng)殖場(chǎng)的家禽疫病非接觸式、快速和自動(dòng)識(shí)別具有重要意義。

[1] Brown Mac, Moore Leslie, Mcmahon Benjamin, et al. Constructing rigorous and broad biosurveillance networks for detecting emerging zoonotic outbreaks[J]. Plos One, 2015, 10(5): e124037.

[2] Mdbazlurr Mollah, Mda Hasan, Mda Salam, et al. Digital image analysis to estimate the live weight of broiler[J]. Computers & Electronics in Agriculture, 2010, 72(1): 48－52.

[3] Aydin A, Bahr C, Viazzi S, et al. A novel method to automatically measure the feed intake of broiler chickens by sound technology[J]. Computers & Electronics in Agriculture, 2014, 101(1): 17－23.

[4] Aydin A, Bahr C, Berckmans D. A real-time monitoring tool to automatically measure the feed intakes of multiple broiler chickens by sound analysis[J]. Computers and Electronics in Agriculture, 2015, 114:1－6.

[5] 浦雪峰，朱偉興，陸晨芳. 基于對(duì)稱像素塊識(shí)別的病豬行為監(jiān)測(cè)系統(tǒng)[J]. 計(jì)算機(jī)工程，2009，35(21)：250－252.

Pu Xuefeng, Zhu Weixing, Lu Chenfang. Sick pig behavior monitor system based on symmetrical pixel block recognization[J]. Computer Engineering, 2009, 35(21): 250－252. (in Chinese with English abstract)

[6] Shao J, Xin H, Harmon J D. Comparison of image feature extraction for classification of swine thermal comfort behavior[J]. Computers and Electronics in Agriculture, 1998, 19(3): 223－232.

[7] Shao B, Xin H. A real-time computer vision assessment and control of thermal comfort for group-housed pigs[J]. Computers and Electronics in Agriculture, 2008, 62(1): 15－21.

[8] 劉龍申，沈明霞，柏廣宇，等. 基于機(jī)器視覺的母豬分娩檢測(cè)方法研究[J]. 農(nóng)業(yè)機(jī)械學(xué)報(bào)，2014，45(3)：237－242.

Liu Longshen, Shen Mingxia, Bo Guangyu, et al. Sows parturition detection method based on machine vision[J]. Transactions of the Chinese Society for Agricultural Machinery, 2014, 45(3): 237－242. (in Chinese with English abstract)

[9] 畢敏娜，張鐵民，莊曉霖，等. 基于色差信息多色彩模型的黃羽雞快速分割方法[J]. 農(nóng)業(yè)機(jī)械學(xué)報(bào)，2016，47(12)：293－298.

Bi Minna, Zhang Tiemin, Zhuang Xiaolin, et al. Fast segmentation method of yellow feather chicken based on difference of color information in different color models[J]. Transactions of the Chinese Society for Agricultural Machinery, 2016, 47(12): 293－298. (in Chinese with English abstract)

[10] 畢敏娜，張鐵民，莊曉霖，等. 基于雞頭特征的病雞識(shí)別方法研究[J]. 農(nóng)業(yè)機(jī)械學(xué)報(bào)，2018，49(1)：51－57.

Bi Minna, Zhang Tiemin, Zhuang Xiaolin, et al. Recognition method of yellow feather chicken based on head features[J]. Transactions of the Chinese Society for Agricultural Machinery, 2018, 49(1): 51－57. (in Chinese with English abstract)

[11] Zhuang Xiaolin, Bi Minna, Guo Jilei, et al. Development of an early warning algorithm to detect sick broilers[J]. Computers & Electronics in Agriculture, 2018, 144: 102－113.

[12] 王琳，孫傳恒，李文勇，等. 基于深度圖像和BP神經(jīng)網(wǎng)絡(luò)的肉雞體質(zhì)量估測(cè)模型[J]. 農(nóng)業(yè)工程學(xué)報(bào)，2017，33(13)：199－205.

Wang Lin, Sun Chuanheng, Li Wenyong, et al. Establishment of broiler quality estimation model based on depth image and BP neural network[J]. Transactions of the Chinese Society of Agricultural Engineering (Transactions of the CSAE), 2017, 33(13): 199－205. (in Chinese with English abstract)

[13] 勞鳳丹，滕光輝，李軍，等. 機(jī)器視覺識(shí)別單只蛋雞行為的方法[J]. 農(nóng)業(yè)工程學(xué)報(bào)，2012，28(24)：157－163.

Lao Fengdan, Teng Guanghui, Li Jun, et al. Behavior recognition method for individual laying hen based on computer vision[J]. Transactions of the Chinese Society of Agricultural Engineering (Transactions of the CSAE), 2012, 28(24): 157－163. (in Chinese with English abstract)

[14] 劉恒，吳迪，蘇家儀，等. 運(yùn)用高斯混合模型識(shí)別動(dòng)物聲音情緒[J]. 國(guó)外電子測(cè)量技術(shù)，2016，35(11)：82－87.

Liu Heng, Wu Di, Su Jiayi, et al. Recognition of animal sound’s emotion based on gaussian mixture model[J]. Foreign Electronic Measurement Technology, 2016, 35(11): 82－87. (in Chinese with English abstract)

[15] 汪開英，趙曉洋，何勇. 畜禽行為及生理信息的無損監(jiān)測(cè)技術(shù)研究進(jìn)展[J]. 農(nóng)業(yè)工程學(xué)報(bào)，2017，33(20)：197－209.

Wang Kaiying, Zhao Xiaoyang, He Yong. Review on no moninavasive monitoring technology of poultry behavior and physiological information[J]. Transactions of the Chinese Society of Agricultural Engineering (Transactions of the CSAE), 2017, 33(20): 197－209. (in Chinese with English abstract)

[16] 于天福. 基于聲音特征的動(dòng)物行為識(shí)別系統(tǒng)研究[D]. 哈爾濱：東北林業(yè)大學(xué)，2010.

Yu Tianfu. The Research of Animal Behavior Recognition System Based on Sound Features[D]. Harbin: Northeast Forestry University, 2010. (in Chinese with English abstract)

[17] 王丹聰. 基于多傳感器融合的豬只行為辨別[D]. 西安：陜西科技大學(xué)，2018.

Wang Dancong. Pig Behavior Identification Based on Multisensor Fusion[D]. Xi’an: Shaanxi University of Science and Technology, 2018. (in Chinese with English abstract)

[18] Moura D J, Silva W T, Naas I A, et al. Real time computer stress monitoring of piglets using vocalization analysis[J]. Computers and Electronics in Agriculture, 2008, 64(1): 11－18.

[19] Moura D J, Silva W T, Naas I A, et al. Real time computer stress monitoring of piglets using vocalization analysis[J]. Computers and Electronics in Agriculture, 2008, 64(1): 11－18.

[20] Guarino M, Jans P, Costa A, et al. Field test of algorithm for automatic cough detection in pig houses[J]. Computers and Electronics in Agriculture, 2008, 62(1): 22－28.

[21] Ferrari S, Piccinini R, Silva M, et al. Cough sound description in relation to respiratory diseases in dairy calves[J]. Preventive Veterinary Medicine, 2010, 96(3/4): 276－280.

[22] Milone D H, Rufiner H L, Galli J R, et al. Computational method for segmentation and classification of ingestive sounds in sheep[J]. Computers and Electronics in Agriculture, 2009, 65(2): 228－237.

[23] Banakar Ahmad, Sadeghi Mohammad, Shushtari Abdolhamid. An intelligent device for diagnosing avian diseases: Newcastle, infectious bronchitis, avian influenza[J]. Computers & Electronics in Agriculture, 2016, 127: 744－753.

[24] 曹晏飛. 復(fù)雜背景下蛋雞聲音分類提取方法研究[D]. 北京：中國(guó)農(nóng)業(yè)大學(xué)，2015.

Cao Yanfei. Research on Methods for Classification and Extraction of Laying Hens' Vocalizations in Complex Environment[D]. Beijing: China Agricultural University, 2015. (in Chinese with English abstract)

[25] 余禮根，滕光輝，李保明，等. 蛋雞發(fā)聲音頻數(shù)據(jù)庫(kù)的構(gòu)建與應(yīng)用[J]. 農(nóng)業(yè)工程學(xué)報(bào)，2012，28(24)：150－156.

Yu Ligen, Teng Guanghui, Li Baoming, et al. Development and application of audio database for laying hens[J]. Transactions of the Chinese Society of Agricultural Engineering (Transactions of the CSAE), 2012, 28(24): 150－156. (in Chinese with English abstract)

[26] Oppenheim A V, Schafer R W, Buck W J R. Discrete-time signal processing[J]. Electronics & Power, 2009, 23(2):157.

[27] Allen J B, Rabiner L. A unified approach to short-time Fourier analysis and synthesis[J]. Proc IEEE, 1977, 65(11): 1558－1564.

[28] Portnoff M. Short-time Fourier analysis of sampled speech[J]. IEEE Transactions on Acoustics Speech & Signal Processing, 1981, 29(3): 364－373.

[29] Immerseel L V, Peeters S. Digital implementation of linear gammatone filters: Comparison of design methods[J]. Acoustics Research Letters Online, 2003, 4(3):59－64.

[30] Quatieri Thomas. Discrete-time Speech Signal Processing: Principles and Practice[M]. Prentice Hall, 2002.

[31] Shannon Claude E, Weaver Warren. The mathematical theory of communication[J]. Computers in Medical Practice, 1950, 3(9): 31－32.

[32] Shen J L, Hung J W, Lee L S. "Robust entropy-based endpoint detection for speech recognition in noisy environments"[C]// The 5th International Conference on Spoken Language Processing, Incorporating the 7th Australian International Speech Science and Technology Conference, Sydney Convention Centre, Sydney, Australia, 1998.

[33] Sun Zengqi, Xu Hongbin. Fuzzy neural network based on T S model[J]. Journal of Tsinghua University, 1997, 37(3):76－80.

[34] Takagi T, Sugeno M. Fuzzy identification of systems and its applications to modeling and control[J]. Readings in Fuzzy Sets for Intelligent Systems, 1993, 15(1): 387－403.

[35] Jin Yaochu, Jiang J, Zhu J.Neural network based fuzzy identification and its application to modeling and control of complex systems[J]. IEEE Transactions on Systems, Man and Cybernetics, 1995, 25(6):990－997.

[36] Takagi Tomohiro, Sugeno Michio. Fuzzy identification of systems and its applications to modeling and control[J]. IEEE Transactions on Systems Man & Cybernetics, 1985, 15(1): 116－132.

Detection of chicken infected with avian influenza based on audio features and fuzzy neural network

Zhang Tiemin1,2,Huang Junduan1

(1.,,510642,; 2.,,510642,)

Avian influenza influences the economy, food safety and human health. A rapid and accurate detection of chicken infected with avian influenza in farming not only directly benefits the chicken farming, but also prevents the cross propagation of avian influenza. This paper proposes a non-invasive disease poultry detection method based on voice analysis, which is designed to achieve the identification of the voice of chickens infected with avian influenza and that of the healthy ones. First, 14 white leghorn chickens of 5 weeks of age with specific pathogen free (SPF) were put into the isolated cage in the animal biosafety level 3 (ABSL 3) laboratory to record their voice. The voice samples of healthy chickens were collected by a T&F-91 enhanced 32G digital HD recording pen, and then the chickens were inoculated with the H7N9 avian influenza virus in the ABSL-3 laboratory. The H7N9 subtype avian influenza virus was diluted to 106EID50/0.1 mL with 10 000/mL penicillin and streptomycin free phosphate-buffered saline (PBS), which was then used to inoculate the chickens, each with 0.1 mL virus diluent. After that, the samples of infected chickens’ voice were collected. Secondly, in light of the fact that the frequency of chickens’ voice signal was higher than the ambient noise, the recorded voice signal was processed with pre-emphasis. The high pass filter was used, so as to weaken the signal of the noise and improve that of chickens’ voice. Thirdly, the processed chicken voice signal was further treated with the hamming window, and then it was divided into smaller segment, 21.3 ms per frames, which could be regarded as quasi steady state process. Fourthly, because the spectral entropy values of the obtained chickens’ voice and the noise were significantly distinguishing, the values of each frame were calculated out. Based on these values, the end point detection method was put forward, so that the chickens’ voice fragments were extracted from the complex ambient noise-containing record, while the non-chicken voice was discarded. Fifthly, the extracted chickens’ voice fragments were treated with time domain analysis, and 3 attributes (short time zero crossing rate, short time energy and the combination of them) were figured out as the characteristics of the healthy chickens and chickens infected with avian influenza. The 450 sampling voice of the healthy chickens and 450 of chicken infected with avian influenza were marked before their order being randomly disrupted. The marked samples were divided into 4 groups: 1 training set (600 samples) and 3 testing sets (100 samples in each group). Finally, the training set was trained by 3 Takagi-Sugeno (T-S) fuzzy neural networks (each with different types of the membership function: π function, Gaussian function and Bell function). It was revealed from the training result that the network with the bell function had the highest recognition rate. So the network with bell shape function was applied to the 3 testing sets and results were obtained respectively: the sensitivity was 75.47%, 80.39% and 76.92%, the specificity was 80.85%, 79.59% and 72.92%, and the true recognition rate was 78%, 80% and 75%. Therefore, this kind of detection method might provide a set of non-invasive, rapid and efficient methods for avian influenza infected chickens detection or identification in poultry farms and poultry circulation market.

neural network; recognition; extraction; spectral entropy;short time zero crossing rate; short time energy; chicken infected with avian influenza detection

10.11975/j.issn.1002-6819.2019.02.022

TP3-05

A

1002-6819(2019)-02-0168-07

2018-07-04

2018-12-30

國(guó)家重點(diǎn)研發(fā)計(jì)劃項(xiàng)目資助（2018YFD0500705）

張鐵民，教授，博士，主要從事智能檢測(cè)與控制研究。Email：tm-zhang@163.com

張鐵民，黃俊端. 基于音頻特征和模糊神經(jīng)網(wǎng)絡(luò)的禽流感病雞檢測(cè)[J]. 農(nóng)業(yè)工程學(xué)報(bào)，2019，35(2)：168－174 doi：10.11975/j.issn.1002-6819.2019.02.022 http://www.tcsae.org

Zhang Tiemin, Huang Junduan. Detection of chicken infected with avian influenza based on audio features and fuzzy neural network[J]. Transactions of the Chinese Society of Agricultural Engineering (Transactions of the CSAE), 2019, 35(2): 168－174. (in Chinese with English abstract) doi：10.11975/j.issn.1002-6819.2019.02.022 http://www.tcsae.org