丁啟朔,董盛盛,李毅念,邱 威,薛金林,何瑞銀(南京農(nóng)業(yè)大學(xué)工學(xué)院, 南京210031)
?
耕層構(gòu)造的土壤結(jié)構(gòu)質(zhì)量-徑級(jí)數(shù)字圖像分析
丁啟朔,董盛盛,李毅念,邱威,薛金林,何瑞銀
(南京農(nóng)業(yè)大學(xué)工學(xué)院, 南京210031)
摘要:為了探討以土壤結(jié)構(gòu)體為單元的耕層構(gòu)造定量方法,該文利用犁耕生成的土壤結(jié)構(gòu)體2D圖像計(jì)算其質(zhì)量-徑級(jí)分布。分別以30°、45°、60°及90°拍攝獲取土壤結(jié)構(gòu)體的數(shù)字圖像,計(jì)算土壤結(jié)構(gòu)體各徑級(jí)區(qū)間的幾何指標(biāo),擬合質(zhì)量-徑級(jí)分布模型。結(jié)果表明土壤結(jié)構(gòu)體的棱角性和形狀指數(shù)隨徑級(jí)增大而增加,但矩形度隨之減小;以60°拍攝所得的土壤結(jié)構(gòu)體質(zhì)量-投影面積關(guān)系的擬合精度最高,各粒徑區(qū)間R2均不低于0.89;數(shù)字圖像篩分與手工測(cè)量所得的土壤結(jié)構(gòu)體質(zhì)量-徑級(jí)分布無顯著差異(P>0.05),表明數(shù)字圖像篩分是從2D投影面信息獲取土壤結(jié)構(gòu)體質(zhì)量-徑級(jí)分布的準(zhǔn)確方法;相對(duì)于Weibull和Rosin-Rammler模型,用Gaudin-Schuhmann模型擬合獲得的土壤結(jié)構(gòu)體質(zhì)量-徑級(jí)分布效果較優(yōu),用該模型擬合數(shù)字圖像篩分所得的土壤結(jié)構(gòu)體質(zhì)量-徑級(jí)分布,R2為0.98;相對(duì)于干篩法,數(shù)字圖像篩分方法的劃分的徑級(jí)區(qū)間更精細(xì),所得的模型擬合精度更高。
關(guān)鍵詞:土壤;模型;圖像處理;耕層構(gòu)造;土壤結(jié)構(gòu)體;數(shù)字圖像;質(zhì)量-徑級(jí)分布
丁啟朔,董盛盛,李毅念,邱威,薛金林,何瑞銀. 耕層構(gòu)造的土壤結(jié)構(gòu)質(zhì)量-徑級(jí)數(shù)字圖像分析[J]. 農(nóng)業(yè)工程學(xué)報(bào),2016,32(2):134-140.doi:10.11975/j.issn.1002-6819.2016.02.020http://www.tcsae.org
Ding Qishuo, Dong Chengsheng, Li Yinian, Qiu Wei, Xue Jinlin, He Ruiyin. Digital image processing of mass-size distribution of soil structures in plough layer[J]. Transactions of the Chinese Society of Agricultural Engineering (Transactions of the CSAE), 2016,32(2): 134-140. (in Chinese with English abstract)doi:10.11975/j.issn.1002-6819.2016.02.020http://www.tcsae.org
合理的耕層構(gòu)造是土地高產(chǎn)出的重要保障,能夠通過調(diào)控土壤的水、肥、氣、熱等環(huán)境要素促進(jìn)農(nóng)作物生長發(fā)育[1-6]。已有針對(duì)耕層構(gòu)造的研究多集中在其肥效、土壤水以及耕層構(gòu)造模式對(duì)作物產(chǎn)量的影響等方面[6-13]。耕層構(gòu)造的“虛、實(shí)”等狀態(tài)多進(jìn)行定性的描述,如全實(shí)耕層、虛實(shí)并存等[1,5-6],耕層構(gòu)造的定量方法多用土壤的硬度、容重、導(dǎo)水率、持水率等指標(biāo)[7,13-18]。然而對(duì)于多數(shù)黏性土壤的耕作而言,耕層構(gòu)造多以土壤結(jié)構(gòu)體的形式存在。以土壤結(jié)構(gòu)體為基本結(jié)構(gòu)單元的耕層構(gòu)造屬性未能得到準(zhǔn)確的界定和定量的描述,耕層的土壤結(jié)構(gòu)體單元及由結(jié)構(gòu)單元磊結(jié)的耕層構(gòu)造定量方法缺乏。
土壤結(jié)構(gòu)體的感官評(píng)價(jià)是土壤物理學(xué)的常規(guī)分析方法[19]。相關(guān)學(xué)者提供了土壤結(jié)構(gòu)體的外觀評(píng)價(jià)指標(biāo)與方法[20-22],但是目前感官評(píng)價(jià)法仍存在主觀性大、統(tǒng)一性差等狀況。相對(duì)于土壤結(jié)構(gòu)體的感官評(píng)價(jià)存在的主觀因素而言,用于土壤團(tuán)聚體及土壤孔隙的定量方法較多,包括土壤切片、計(jì)算機(jī)斷層掃描(computer tomography,CT)和核磁共振波譜法(nuclear magnetic resonance spectroscopy, NMR)[23-25]。然而這些用于土壤結(jié)構(gòu)分析的定量方法主要涉及土壤的微形態(tài)特征分析[26],并不適用于耕層構(gòu)造的土壤結(jié)構(gòu)體定量分析。
現(xiàn)有的耕層構(gòu)造土壤結(jié)構(gòu)體定量方法及相關(guān)國家標(biāo)準(zhǔn)使用篩分操作[27-28],所得到的數(shù)據(jù)是土壤結(jié)構(gòu)體的徑級(jí)-質(zhì)量分布信息。然而用于原位快速測(cè)試的數(shù)字圖像分析方法通常提供土壤結(jié)構(gòu)體的2D平面信息[29],因此需要探討使用2D信息獲取土壤結(jié)構(gòu)體質(zhì)量信息的方法。本文在高雅等[29]工作的基礎(chǔ)上,研究從土壤結(jié)構(gòu)體2D數(shù)字圖像轉(zhuǎn)換為土壤結(jié)構(gòu)體質(zhì)量信息方法,與此同時(shí),推斷耕后土壤結(jié)構(gòu)體的徑級(jí)分布,并結(jié)合土壤結(jié)構(gòu)體的外觀幾何指標(biāo)描述水稻土犁耕處理后耕層構(gòu)造及土壤結(jié)構(gòu)體的狀態(tài)信息。
1.1土壤樣本的獲取方法
田間試驗(yàn)于2014年11月19日水稻收獲后進(jìn)行,試驗(yàn)地點(diǎn)位于南京市浦口區(qū)江浦農(nóng)場(chǎng)(118°59′E,31°98′N),土壤類型是黏性水稻土,該區(qū)常年稻麥輪作。土壤pH值7.6,土壤砂粒、壤粒、黏粒、有機(jī)物質(zhì)量分?jǐn)?shù)分別為21.3%、39.84%、38.85%和3.18%,液限47.33%,塑限26.67%。
使用田間原位綜合耕作試驗(yàn)臺(tái)[30]掛接鏵式犁進(jìn)行測(cè)試,單鏵犁工作幅寬21 cm,控制耕深為15 cm,牽引速度為0.2 m/s,耕作測(cè)試的有效行程3 m,試驗(yàn)過程中測(cè)取0~15 cm土層的含水率、容重分別為30.79%、1.61 mg/m3,試驗(yàn)隨機(jī)重復(fù)3次。原位自然風(fēng)干2天后緩慢地取出耕作范圍內(nèi)大徑級(jí)土垡并放置在海綿墊上,同時(shí)集取耕作范圍內(nèi)的較小的土壤結(jié)構(gòu)體。所得土壤結(jié)構(gòu)體粒徑分主要布于4~180 mm范圍內(nèi),其中小于4 mm粒徑的土壤結(jié)構(gòu)體質(zhì)量占總質(zhì)量6.6%,3次試驗(yàn)土壤結(jié)構(gòu)體粒徑在128~180 mm數(shù)量較少,分別為6個(gè)、7個(gè)、7個(gè)。
1.2 土壤結(jié)構(gòu)體的圖像采集與篩分
土壤結(jié)構(gòu)體的數(shù)字圖像采集、標(biāo)定及預(yù)處理參照高雅等[29]的方法,圖像采集設(shè)備如圖1所示。將土壤結(jié)構(gòu)體隨機(jī)平鋪于尺度為80 cm×80 cm的淺色背景板上,三腳支架上安裝的數(shù)碼相機(jī)型號(hào)為Cannon A 3300,拍攝時(shí)光圈焦距調(diào)至F3.2。俯視角拍攝時(shí),相機(jī)與土壤結(jié)構(gòu)體垂直距離為1 m,與背景板中心水平距離為80 cm(圖1);90°拍攝時(shí),相機(jī)位于背景板中點(diǎn)正上方,與土壤結(jié)構(gòu)體垂直距離為1 m。所有拍攝的圖片以JPEG格式保存。
數(shù)字圖像采集完成后,對(duì)3次試驗(yàn)樣本分別進(jìn)行手工測(cè)量,采用2倍頻徑級(jí)劃分區(qū)間,為了較少地破壞土壤結(jié)構(gòu)體,手工測(cè)量使用不同徑級(jí)機(jī)加工的圓環(huán)逐個(gè)測(cè)量土壤結(jié)構(gòu)體,記錄其質(zhì)量與相應(yīng)徑級(jí)。手工測(cè)量與干篩法采用2倍頻徑級(jí)劃分區(qū)間,徑級(jí)區(qū)間為>4~16 mm、>16~32 mm、>32~64 mm、>64~128mm、>128 mm,在試驗(yàn)取樣中發(fā)現(xiàn)>128 mm土壤結(jié)構(gòu)體數(shù)量占樣本總量的6‰,將>64~128mm、>128 mm土壤結(jié)構(gòu)體劃分至>64 mm徑級(jí)區(qū)間。
圖1 圖像采集裝置Fig.1 Image acquisition device
1.3土壤結(jié)構(gòu)體的數(shù)字圖像篩分及徑級(jí)分布
數(shù)字圖像分析方法已應(yīng)用于土壤結(jié)構(gòu)數(shù)量-徑級(jí)分布[29],但從2D圖像計(jì)算土壤結(jié)構(gòu)體的質(zhì)量信息尚沒有系統(tǒng)研究,而且拍攝2D圖像的角度不同轉(zhuǎn)換得到的質(zhì)量信息也不相同,因此在預(yù)備試驗(yàn)中分別按照俯視角0°、15°、30°、45°、60°、75°及90°對(duì)土壤結(jié)構(gòu)體進(jìn)行拍攝,計(jì)算單個(gè)土壤結(jié)構(gòu)體質(zhì)量與該土壤結(jié)構(gòu)體圖像面積的對(duì)應(yīng)關(guān)系??紤]到田間耕作的在線檢測(cè)過程較難使用小角度拍攝(0°及15°),而且預(yù)備試驗(yàn)也發(fā)現(xiàn)75°拍攝得到的結(jié)果與90°拍攝非常相近,因此,本文研究時(shí)僅使用30°、45°、60°及90°的拍攝結(jié)果進(jìn)行對(duì)比。
本文使用MATLAB中NLIN(非線性)程序擬合得到土壤結(jié)構(gòu)體質(zhì)量-徑級(jí)分布統(tǒng)計(jì)模型[31-32],并將干篩法得到的土壤結(jié)構(gòu)體質(zhì)量-徑級(jí)及數(shù)量-徑級(jí)分布信息與數(shù)字圖像分析的結(jié)果進(jìn)行統(tǒng)計(jì)學(xué)分析。鑒于土壤結(jié)構(gòu)體質(zhì)量-徑級(jí)分布狀態(tài)的描述多用模型表達(dá),因此本文選擇3種模型(Weibull模型[33]、Rosin-Rammler模型[34]、Gaudin-Schuhmann模型[34])描述犁耕后水稻土耕層構(gòu)造的土壤結(jié)構(gòu)體質(zhì)量—徑級(jí)分布,3種模型的方程列于表1。
表1 土壤結(jié)構(gòu)體質(zhì)量—徑級(jí)分布模型及參數(shù)Table 1 Soil structure mass-size distribution models with parameters
2.1土壤結(jié)構(gòu)體外觀幾何指標(biāo)
數(shù)字圖像分析能夠按照任意的徑級(jí)區(qū)間對(duì)土壤結(jié)構(gòu)體進(jìn)行圖像篩分[29],按8 mm等間距徑級(jí)劃分得到土壤結(jié)構(gòu)體的棱角性、矩形度、形狀指數(shù)隨徑級(jí)變化規(guī)律如圖2所示,隨徑級(jí)的增大,棱角性呈線性遞增趨勢(shì),矩形度隨徑級(jí)增加有下降的趨勢(shì)。犁耕處理的耕層構(gòu)造土壤結(jié)構(gòu)體在不同的徑級(jí)范圍具有不同的外觀幾何特性,在一定程度上,數(shù)字圖像分析作為一個(gè)精細(xì)的定量方法能夠?yàn)楦麑訕?gòu)造的土壤結(jié)構(gòu)體單元提供幾何外觀的模型信息。
圖2 土壤結(jié)構(gòu)體幾何指標(biāo)Fig.2 Box plots of soil structure geometry indicator
2.2土壤結(jié)構(gòu)體2D投影面積與質(zhì)量的對(duì)應(yīng)關(guān)系
將俯視角30°、45°、60°以及90°的拍攝結(jié)果進(jìn)行處理,所得土壤結(jié)構(gòu)體的2D投影面積與對(duì)應(yīng)的土壤結(jié)構(gòu)體質(zhì)量進(jìn)行擬合,發(fā)現(xiàn)在不同徑級(jí)區(qū)間以及用不同拍攝角所得的土壤結(jié)構(gòu)體投影面積與質(zhì)量的相關(guān)性各不相同。在>4~16 mm徑級(jí)區(qū)間徑級(jí)區(qū)間R2表現(xiàn)為,R2(30°)<R2(90°)<R2(45°)<R2(60°),60°擬合效果最好,其R2為0.89。R2在>16~32 mm以及>32~64 mm表現(xiàn)一致,為R2(90°)<R2(30°)<R2(45°)<R2(60°),60°擬合效果較好,R2分別為0.91 與0.94,在90°拍攝擬合效果很差,可能是由于在該區(qū)間的土壤結(jié)構(gòu)體形狀極其不規(guī)則導(dǎo)致的,需要在今后的研究中進(jìn)一步試驗(yàn)探討;在>64 mm徑級(jí)區(qū)間R2表現(xiàn)為,R2(90°)<R2(60°)<R2(45°)<R2(60°),60°擬合效果較好, R2為0.95。相對(duì)于試驗(yàn)其余拍攝角度,60°拍攝時(shí),土壤結(jié)構(gòu)體質(zhì)量與投影面積擬合精度最優(yōu)。與此同時(shí),隨著土壤結(jié)構(gòu)體徑級(jí)增大,R2逐漸增大。因此為了較為準(zhǔn)確描述土壤結(jié)構(gòu)體質(zhì)量-徑級(jí)分布,本文選擇在俯視角為60°拍攝時(shí)的土壤結(jié)構(gòu)體投影面積與質(zhì)量的函數(shù)關(guān)系。
不同徑級(jí)區(qū)間土壤結(jié)構(gòu)體投影面積(s)與質(zhì)量(m)對(duì)應(yīng)的關(guān)系如圖3所示,>4~16 mm徑級(jí)區(qū)間:m=0.8562s?0.199;>16~32 mm徑級(jí)區(qū)間:m=1.7446s?3.4554;>32~64 mm徑級(jí)區(qū)間:m=3.5626s?20.425;>64 mm徑級(jí)區(qū)間:m=7.5087s? 279.12。然而結(jié)果也進(jìn)一步表明構(gòu)成犁耕耕層構(gòu)造的土壤結(jié)構(gòu)體在不同徑級(jí)范圍的質(zhì)量分布特征并不完全相同。
圖3 60°拍攝各徑級(jí)區(qū)間的土壤結(jié)構(gòu)體質(zhì)量-投影面關(guān)系Fig.3 Mass-projection area relation of soil structures in each size range for 60° projection
2.3土壤結(jié)構(gòu)體徑級(jí)分布
2.3.1土壤結(jié)構(gòu)體質(zhì)量-徑級(jí)分布模型
構(gòu)成犁耕耕層構(gòu)造的土壤結(jié)構(gòu)體徑級(jí)分布模型分別使用Weibull模型、Rosin-Rammler模型和Gaudin-Schuhmann模型進(jìn)行擬合。為了確定土壤結(jié)構(gòu)體質(zhì)量-徑級(jí)分布模型的適用性,分別使用3種評(píng)價(jià)指標(biāo):相關(guān)系數(shù)(correlation coefficient, R2)、均方根誤差(root-meansquare error, RMSE)以及Akaike信息準(zhǔn)則(akaike information criterion, AIC)評(píng)價(jià)模型的擬合精度[35]。其中R2值越大表示擬合效果越好,RMSE和AIC越小擬合效果越好。通過對(duì)3種模型擬合精度分析表明(表2),對(duì)于R2,Gaudin-Schuhmann模型>W(wǎng)eibull模型> Rosin-Rammler模型,Gaudin-Schuhmann模型與Rosin-Rammler模型差異顯著(P<0.05),與Weibull模型差異不顯著(P>0.05),Gaudin-Schuhmann模型R2為0.93;RMSE和AIC表現(xiàn)一致,均為Gaudin-Schuhmann模型<Weibull模型<Rosin-Rammler模型,Gaudin-Schuhmann模型與Weibull模型、Rosin-Rammler模型差異顯著(P<0.05)。綜合比較得出Gaudin-Schuhmann模型的擬合效果最好,Weibull模型次之,Rosin-Rammler 模型擬合效果最差。因此用Gaudin-Schuhmann模型表達(dá)水稻土耕層構(gòu)造的土壤結(jié)構(gòu)體質(zhì)量-徑級(jí)分布效果最優(yōu)。
表2 不同模型的相關(guān)系數(shù)、均方根誤差和信息準(zhǔn)則Table 2 Correlation coefficient, Root-mean-square error and Akaike information criterion of different models
2.3.2土壤結(jié)構(gòu)體的數(shù)字圖像篩分
將手工測(cè)量所得土壤結(jié)構(gòu)體的質(zhì)量-徑級(jí)及數(shù)量徑級(jí)分布信息與數(shù)字圖像篩分的結(jié)果進(jìn)行統(tǒng)計(jì)分析(表3)。在<8 mm徑級(jí)區(qū)間內(nèi)2種方法所得的土壤結(jié)構(gòu)體質(zhì)量-徑級(jí)分布(累積質(zhì)量分?jǐn)?shù))差異顯著(P<0.05),而在<16 mm、<32 mm、<64 mm、<128 mm 以及<256 mm徑級(jí)區(qū)間,手工測(cè)量與數(shù)字圖像篩分的結(jié)果差異不顯著(P>0.05)。表明8 mm以下的土壤結(jié)構(gòu)體粒徑較小,圖像分析的誤差相對(duì)較大,而8 mm以上土壤結(jié)構(gòu)體的圖像分析結(jié)果較為準(zhǔn)確。土壤結(jié)構(gòu)體的質(zhì)量-徑級(jí)分布的手工測(cè)量與數(shù)字圖像結(jié)果差異不顯著(P>0.05),說明用數(shù)字圖像分析方法得到土壤結(jié)構(gòu)體的2D平面信息能夠準(zhǔn)確描述土壤結(jié)構(gòu)體3D(質(zhì)量)信息。
表3 2種方法的土壤結(jié)構(gòu)體—徑級(jí)分布Table 3 Cumulative percentage soil structure distribution of two methods
使用3種土壤結(jié)構(gòu)體質(zhì)量-徑級(jí)分布模型中最優(yōu)的Gaudin-Schuhmann模型擬合(表2),干篩法以2倍頻獲得的土壤結(jié)構(gòu)體質(zhì)量-徑級(jí)分布模型為F( d< D)=( d/252.102)0.77,R2為0.93(圖4a);數(shù)字圖像篩分以4 mm等間距徑級(jí)分布獲得的土壤結(jié)構(gòu)體的質(zhì)量-徑級(jí)分布模型為F( d< D)=( d /171)1.096,R2為0.98(圖4b)。對(duì)于土壤結(jié)構(gòu)體的數(shù)量徑級(jí)分布而言,干篩法所得模型為w1=116.95ln(d)+653.87,R2為0.74(圖4c);數(shù)字圖像篩分得到的模型為w2=136.93ln(d)+582.8,R2為0.83(圖4d)。相對(duì)于干篩法,數(shù)字圖像篩分能夠便捷、精細(xì)的劃分區(qū)間,所得質(zhì)量-徑級(jí)分布模型與數(shù)量徑級(jí)分布模型精度更高。
使用模型描述土壤結(jié)構(gòu)體的質(zhì)量-徑級(jí)分布的優(yōu)點(diǎn)是土壤結(jié)構(gòu)體的形狀參數(shù)能夠量化。干篩法和數(shù)字圖像篩分所得的模型的形狀參數(shù)λ分別為252.102和171,造成這一差異的原因是手工篩分所用的篩孔徑級(jí)為2倍頻徑級(jí),間距較大,徑級(jí)的上限為256 mm。與此不同,數(shù)字圖像篩分按照取樣所得的實(shí)際土壤結(jié)構(gòu)體粒徑最大值確定(最大粒徑值為180 mm),因此可見模型的參數(shù)λ與土壤結(jié)構(gòu)體的徑級(jí)上限相關(guān),Diaz-Zorita[34]等的研究中也得到同樣的結(jié)論。土壤結(jié)構(gòu)的質(zhì)量-徑級(jí)分布模型將耕后土壤結(jié)構(gòu)參量化,從而有助于建立更高階的土壤耕作力學(xué)模型[28]。試驗(yàn)發(fā)現(xiàn)>32 mm土壤結(jié)構(gòu)體的質(zhì)量份額超過樣本總質(zhì)量的60%(圖4c,d),表明水稻土條件下犁耕的耕層構(gòu)造質(zhì)量較差,碎土效果不良。Arvidsson[36]研究發(fā)現(xiàn),在黏性土壤條件下犁耕的耕層構(gòu)造質(zhì)量較差,土壤結(jié)構(gòu)體(>32 mm)質(zhì)量分?jǐn)?shù)為47%。
圖4 土壤結(jié)構(gòu)體徑級(jí)分布Fig.4 Soil structure size distribution
耕層構(gòu)造是耕層土壤的三相搭配及壘結(jié)[37],土壤結(jié)構(gòu)體(土垡、土塊)由機(jī)械耕作生成,是構(gòu)成耕層構(gòu)造的基本結(jié)構(gòu)單元。對(duì)耕層構(gòu)造的土壤結(jié)構(gòu)的研究主要采用土壤團(tuán)聚體的組成以及穩(wěn)定性等指標(biāo)[38]。多數(shù)研究認(rèn)為土壤結(jié)構(gòu)的基本單元是土壤團(tuán)聚體[26,39-40],因此土壤團(tuán)聚體的形成和變化過程、團(tuán)聚體穩(wěn)定性以及影響因素等是土壤結(jié)構(gòu)研究的主要內(nèi)容,其中干篩法、濕篩法是土壤學(xué)中定量描述土壤團(tuán)聚體的常用方法[26,41-42]。在試驗(yàn)預(yù)處理中,人們通常將大土塊按自然裂痕手工剝離為1 cm3左右的土壤結(jié)構(gòu)體并進(jìn)一步篩分[11,41,43],但是這樣獲得的團(tuán)聚體顯然不是機(jī)械耕作創(chuàng)建的(塊狀)土壤結(jié)構(gòu)體。丁啟朔[37]對(duì)機(jī)械耕作生成的土壤結(jié)構(gòu)體徑級(jí)進(jìn)行了界定,耕層構(gòu)造的徑級(jí)范疇在cm-dm水平[37,44],高出土壤結(jié)構(gòu)一個(gè)徑級(jí)量級(jí)以上,可見現(xiàn)有的針對(duì)耕層構(gòu)造的土壤結(jié)構(gòu)體定量方法存在欠缺,限制了人們對(duì)耕層構(gòu)造的定量描述。
機(jī)械耕作的耕層構(gòu)造多為土壤結(jié)構(gòu)體的磊結(jié)體,高雅[29]等的研究中,提出了棱角性、矩形度、形狀指數(shù)3個(gè)幾何指標(biāo),但是沒有在特定的土壤耕作條件下推廣應(yīng)用,本文將3個(gè)指標(biāo)應(yīng)用于水稻土犁耕的土壤結(jié)構(gòu)體,反映了田間水稻土的真實(shí)的幾何特征。周虎[24]認(rèn)為:土壤結(jié)構(gòu)的定量研究一直受到方法和理論的困擾,常規(guī)的描述土壤結(jié)構(gòu)體的指標(biāo)是篩分指標(biāo),體現(xiàn)的是‘質(zhì)量-徑級(jí)’信息。孔令德[45]研究表明旋耕后的塊狀土壤結(jié)構(gòu)體徑級(jí)分布符合分形特征。Fernlund[46]提出3D圖像分析方法的必要性,通過獲取兩個(gè)互相垂直投影面的圖像分析土壤結(jié)構(gòu)體徑級(jí)徑級(jí)分布,但該方法繁瑣且難于消除水平拍攝的誤差。高雅等[29]的研究中,體現(xiàn)了“數(shù)量-徑級(jí)”信息,沒有對(duì)“質(zhì)量-徑級(jí)”信息深入研究。本文使用數(shù)字圖像分析技術(shù),通過不同角度拍攝的土壤結(jié)構(gòu)體的2D圖像信息獲取土壤結(jié)構(gòu)體質(zhì)量-徑級(jí)分布,對(duì)比手工測(cè)量與數(shù)字圖像篩分所得土壤結(jié)構(gòu)體質(zhì)量-徑級(jí)分布,表明數(shù)字圖像分析方法能夠較為準(zhǔn)確得描述土壤結(jié)構(gòu)體質(zhì)量-徑級(jí)分布。
耕作機(jī)械的國家和行業(yè)標(biāo)準(zhǔn)[27-28]以及人們?cè)谘邪l(fā)新型耕作機(jī)械過程中對(duì)碎土質(zhì)量的測(cè)定和評(píng)價(jià)都使用碎土率[47-48]指標(biāo),然而碎土率指標(biāo)既不能反映土塊的徑級(jí)分布也無法定量土塊的幾何外觀特征,從而導(dǎo)致人們?cè)诿枋龈麑訕?gòu)造時(shí)缺少相應(yīng)的定量手段。土壤結(jié)構(gòu)數(shù)字圖像分析是準(zhǔn)確評(píng)價(jià)耕作機(jī)具的作業(yè)性能的保障,進(jìn)一步的系統(tǒng)開發(fā)可以實(shí)現(xiàn)田間無損在線檢測(cè),如耕層構(gòu)造的實(shí)時(shí)原位在線檢測(cè)[49],不過該技術(shù)的系統(tǒng)集成尚需進(jìn)一步完善。
1)犁耕處理的耕層構(gòu)造土壤結(jié)構(gòu)體在不同的徑級(jí)范圍具有不同的外觀幾何特性,土壤結(jié)構(gòu)體的棱角性和形狀指數(shù)隨徑級(jí)增大有增大的趨勢(shì),而矩形度隨徑級(jí)有減小的趨勢(shì)。
2)犁耕耕層構(gòu)造的土壤結(jié)構(gòu)體在不同徑級(jí)范圍的質(zhì)量分布特征并不完全相同。60°拍攝獲得的土壤結(jié)構(gòu)體質(zhì)量-投影面積的擬合精度較之其他拍攝角度更高;通過比較數(shù)字圖像篩分與手工測(cè)量得到的土壤結(jié)構(gòu)體的質(zhì)量-徑級(jí)分布,在<8 mm徑級(jí)區(qū)間內(nèi)2種方法所得的土壤結(jié)構(gòu)體質(zhì)量-徑級(jí)分布差異顯著(P<0.05),而在<16 mm、<32 mm、<64 mm、<128 mm以及<256 mm徑級(jí)區(qū)間,手工測(cè)量與數(shù)字圖像篩分的結(jié)果差異不顯著(P>0.05)。
3)相對(duì)于Weibull和Rosin-Rammler模型,用Gaudin-Schuhmann模型擬合獲得水稻土條件下犁耕的土壤結(jié)構(gòu)體質(zhì)量-徑級(jí)分布精度更高,用該模型擬合數(shù)字圖像篩分所得的土壤結(jié)構(gòu)體質(zhì)量-徑級(jí)分布,R2為0.98。
[參考文獻(xiàn)]
[1] 白偉,孫占祥,鄭家明,等. 虛實(shí)并存耕層提高春玉米產(chǎn)量和水分利用效率[J]. 農(nóng)業(yè)工程學(xué)報(bào),2014,30(21):81-90. Bai Wei, Sun Zhanxiang, Zheng Jiaming, et al. Furrow loose and ridge compaction plough layer improves spring maize yield and water use efficiency[J]. Transactions of the Chinese Society of Agricultural Engineering (Transactions of the CSAE), 2014, 30(21): 81-90. (in Chinese with English abstract)
[2] 王玉玲,李軍,柏?zé)樝? 輪耕體系對(duì)黃土臺(tái)塬麥玉輪作土壤生產(chǎn)性能的影響[J]. 農(nóng)業(yè)工程學(xué)報(bào),2015,31(1):107-116. Wang Yuling, Li Jun, Bai Weixia. Effects of rotational tillage systems on soil production performance in wheat-maize rotation field in Loess Platform region of China[J]. Transactions of the Chinese Society of Agricultural Engineering (Transactions of the CSAE), 2015, 31(1): 107-116. (in Chinese with English abstract)
[3] 丁昆侖,Hann M J. 耕作措施對(duì)土壤特性及作物產(chǎn)量的影響[J]. 農(nóng)業(yè)工程學(xué)報(bào),2000,16(3):28-31. Ding Kunlun, Hann M J. Effects of soil management on soil properties and crop yield[J]. Transactions of the Chinese Society of Agricultural Engineering (Transactions of the CSAE), 2000, 16(3): 28-31. (in Chinese with English abstract)
[4] 羅洋,鄭金玉,鄭洪兵,等. 不同耕層結(jié)構(gòu)對(duì)玉米生長發(fā)育的影響[J]. 玉米科學(xué),2014,22(2):113-116. Luo Yang, Zheng Jinyu, Zheng Hongbing. et al. Effect on growth of maize in different soil plough layer by tillage[J]. Journal of Maize Sciences, 2014, 22(2): 113-116. (in Chinese with English abstract)
[5] 王立春,馬虹,鄭金玉. 東北春玉米耕地合理耕層構(gòu)造研究[J]. 玉米科學(xué),2008,16(4):13-17. Wang Lichun, Ma Hong, Zheng Jinyu. Research on rational plough layer con struction of spring maize soil in Northeast China[J]. Journal of Maize Sciences, 2008, 16(4): 13-17. (in Chinese with English abstract)
[6] 劉武仁,鄭金玉,羅洋,等. 不同耕層構(gòu)造對(duì)土壤硬度和含水量的影響[J]. 玉米科學(xué),2013,21(6):76-80. Liu Wuren, Zheng Jinyu, Luo Yang, et al. Effects of different tillage layer structures on soil compaction and soil water content[J]. Journal of Maize Sciences, 2013, 21(6): 76-80. (in Chinese with English abstract)
[7] 孔曉民,韓成衛(wèi),曾蘇明,等. 不同耕作方式對(duì)土壤物理性狀及玉米產(chǎn)量的影響[J]. 玉米科學(xué),2014,22(1):38-43. Kong Xiaomin, Han Chengwei, Zeng Suming, et al. Effects of different tillage managements on soil physical properties and maize yield[J]. Journal of Maize Sciences, 2014, 22(1): 38-43. (in Chinese with English abstract)
[8] 秦紅靈,高旺盛,馬月存,等. 兩年免耕后深松對(duì)土壤水分的影響[J]. 中國農(nóng)業(yè)科學(xué),2008,41(1):78-85. Qin Honglin, Gao Wangsheng, Ma Yuecun, et al. Effects of subsoiling on soil moisture under no-tillage 2 years later[J]. Scientia Agricultura Sinica, 2008, 41(1): 78-85. (in Chinese with English abstract)
[9] 黃健,張惠琳,傅文玉,等. 東北黑土區(qū)土壤肥力變化特征的分析[J]. 土壤通報(bào),2005,36(5):659-663. Huang Jian, Zhang Huilin, Fu WenYu, et al. Analysis of the changing characteristics of the soil fertilities in the black soil area of northeast China[J]. Chinese Journal of Soil Science,2005, 36(5): 659-663. (in Chinese with English abstract)
[10] 聶良鵬,郭利偉,牛海燕,等. 輪耕對(duì)小麥-玉米兩熟農(nóng)田耕層構(gòu)造及作物產(chǎn)量與品質(zhì)的影響[J]. 作物學(xué)報(bào),2015,41(3):468-478. Nie Liangpeng, Guo Liwei, Niu Haiyan, et al. Effects of rotational tillage on tilth soil structure and crop yield and quality in maize-wheat cropping system[J]. Acta agronomica sinica 2015, 41(3): 468-478. (in Chinese with English abstract)
[11] 侯賢清,賈志寬,韓清芳,等. 不同輪耕模式對(duì)旱地土壤結(jié)構(gòu)及入滲蓄水特性的影響[J]. 農(nóng)業(yè)工程學(xué)報(bào),2012,28(5):85-94. Hou Xianqing, Jia Zhikuan, Han Qingfang, et al. Effects of different rotational tillage patterns on soil structure,infiltration and water storage characteristics in dryland[J]. Transactions of the Chinese Society of Agricultural Engineering (Transactions of the CSAE), 2012, 28(5): 85-94. (in Chinese with English abstract)
[12] 張銀平,杜瑞成,刁培松,等. 機(jī)械化生態(tài)沃土耕作模式提高土壤質(zhì)量及作物產(chǎn)量[J]. 農(nóng)業(yè)工程學(xué)報(bào),2015,31(7):33-38. Zhang Yinping, Du Ruicheng, Diao Peisong, et al. Mechanical and ecological tillage pattern improving soil quality and crop yields[J]. Transactions of the Chinese Society of Agricultural Engineering (Transactions of the CSAE), 2015, 31(7): 33-38. (in Chinese with English abstract)
[13] 何進(jìn),李洪文,高煥文. 中國北方保護(hù)性耕作條件下深松效應(yīng)與經(jīng)濟(jì)效益研究[J]. 農(nóng)業(yè)工程學(xué)報(bào),2006,22(10):62-67. He Jin, Li Hongwen, Gao Huanwen. Subsoiling effect and economic benefit under conservation tillage mode in Northern China[J]. Transactions of the CSAE, 2006, 22(10): 62-67. (in Chinese with English abstract)
[14] 張祥彩,李洪文,何進(jìn),等. 耕作方式對(duì)華北一年兩熟區(qū)土壤及作物特性的影響[J]. 農(nóng)業(yè)機(jī)械學(xué)報(bào),2013,44(z1). Zhang Xiangcai, Li Hongwen, He Jin, et al. Effects of different tillage managements on characteristics of soil and crop in annual double cropping areas in northern China[J]. Transactions of the CSAE, 2013, 44(z1). (in Chinese with English abstract)
[15] 鄭洪兵,鄭金玉,羅洋,等. 長期不同耕作方式下的土壤硬度變化特征[J]. 農(nóng)業(yè)工程學(xué)報(bào),2015,31(9):63-70. Zheng Hongbing, Zheng Jinyu, Luo Yang, et al. Change characteristic of soil compaction of long-term different tillage methods in cropland[J]. Transactions of the Chinese Society of Agricultural Engineering (Transactions of the CSAE),2015, 31(9): 63-70. (in Chinese with English abstract)
[16] 王巖,劉玉華,張立峰,等. 耕作方式對(duì)冀西北栗鈣土土壤物理性狀及莜麥生長的影響[J]. 農(nóng)業(yè)工程學(xué)報(bào),2014,30(4):109-117. Wang Yan, Liu Yuhua, Zhang Lifeng, et al. Effects of tillage mode on chestnut soil’s physical characters and naked oats growth in Northwest Hebei province[J]. Transactions of the Chinese Society of Agricultural Engineering (Transactions of the CSAE), 2014, 30(4): 109-117. (in Chinese with English abstract)
[17] 張玉嬌,李軍,郭正,等. 渭北旱塬麥田保護(hù)性輪耕方式的產(chǎn)量和土壤水分效應(yīng)長周期模擬研究[J]. 中國農(nóng)業(yè)科學(xué),2015,48(14):2730-2746. Zhang Yujiao, Li Jun, Guo Zheng, et al. Long-term simulation of winter wheat yield and soil water response to conservation tillage rotation in Weibei Highland[J]. Scientia Agricultura Sinica, 2015, 48(14): 2730-2746. (in Chinese with English abstract)
[18] 武均,蔡立群,羅珠珠,等. 保護(hù)性耕作對(duì)隴中黃土高原雨養(yǎng)農(nóng)田土壤物理性狀的影響[J]. 水土保持學(xué)報(bào),2014,28(2):112-117. Wu Jun, Cai Liqun, Luo Zhuzhu, et al. Effects of conservation tillage on soil physical properties of rainfed field of the loess plateau in central of gansu[J]. Journal of Soil & Water Conservation, 2014, 28(2): 112-117. (in Chinese with English abstract)
[19] Askari M S, Cui J, Holden N M. The visual evaluation of soil structure under arable management[J]. Soil & Tillage Research, 2013, 134(8): 1-10.
[20] Mueller L, Shepherd G, Schindler U, et al. Evaluation of soil structure in the framework of an overall soil quality rating[J]. Soil & Tillage Research, 2013, 127(1): 74-84.
[21] Ball B C, Batey T, Munkholm L J. Field assessment of soil structural quality-a development of the Peerlkamp test[J]. Soil Use & Management, 2007, 23(4): 329-337.
[22] Müller L. Soil Structure, Visual Assessment[M]. Springer Netherlands, Encyclopedia of Agrophysics, 2011: 783-786.
[23] 程亞南,劉建立,呂菲,等. 基于CT圖像的土壤孔隙結(jié)構(gòu)三維重建及水力學(xué)性質(zhì)預(yù)測(cè)[J]. 農(nóng)業(yè)工程學(xué)報(bào),2012,28(22):115-122. Cheng Yanan, Liu Jianli, Lü Fei, et al. Three-dimensional reconstruction of soil pore structure and prediction of soil hydraulic properties based on CT images[J]. Transactions of the Chinese Society of Agricultural Engineering (Transactions of the CSAE), 2012, 28(22): 115-122. (in Chinese with English abstract)
[24] 周虎,彭新華,張中彬,等. 基于同步輻射微CT研究不同利用年限水稻土團(tuán)聚體微結(jié)構(gòu)特征[J]. 農(nóng)業(yè)工程學(xué)報(bào),2011,27(12):343-347. Zhou Hu, Peng Xinhua, Zhang Zhongbin, et al. Characterization of aggregate microstructure of paddy soils cultivated for different years with synchrotron based micro-CT[J]. Transactions of the Chinese Society of Agricultural Engineering (Transactions of the CSAE), 27(12): 343-347. (in Chinese with English abstract)
[25] 楊永輝,武繼承,毛永萍,等. 利用計(jì)算機(jī)斷層掃描技術(shù)研究土壤改良措施下土壤孔隙[J]. 農(nóng)業(yè)工程學(xué)報(bào),2013,29(23):99-108. Yang Yonghui, Wu Jicheng, Mao Yongping, et al. Using computed tomography scanning to study soil pores under different soil structure improvement measures[J]. Transactions of the Chinese Society of Agricultural Engineering (Transactions of the CSAE), 2013, 29(23): 99-108. (in Chinese with English abstract)
[26] 畢利東,張斌,潘繼花. 運(yùn)用Image J軟件分析土壤結(jié)構(gòu)特征[J]. 土壤,2009,41(4):654-658. Bi Lidong, Zhang Bin, Pan Jihua. Analysis of Soil Structural Properties by Using Image-J Software[J]. Soils, 2009, 41(4): 654-658. (in Chinese with English abstract)
[27] 中華人民共和國農(nóng)業(yè)部. 深松、耙茬機(jī)械作業(yè)質(zhì)量標(biāo)準(zhǔn):NY-T 741-2003 [B]. 農(nóng)業(yè)部農(nóng)業(yè)機(jī)械化管理司,2003:6 -9.
[28] 中國機(jī)械工業(yè)部.旋耕機(jī)械試驗(yàn)方法:GB/T5668.3-1995 [S].北京:中國標(biāo)準(zhǔn)出版社,1995.
[29] 高雅,丁啟朔,李毅念,等. 土壤結(jié)構(gòu)的數(shù)字圖像分析方法與指標(biāo)[J]. 土壤通報(bào),2015,46(3):513-518. Gao Ya, Ding Qishuo, Li Yinian, et al. A digital image processing tool and indices for soil structure[J]. Chinese Journal of Soil Science, 2015, 46(3): 513-518. (in Chinese with English abstract)
[30] 楊艷山,丁啟朔,丁為民,等. 田間原位綜合耕作試驗(yàn)臺(tái)設(shè)計(jì)與應(yīng)用[J]. 農(nóng)業(yè)機(jī)械學(xué)報(bào),2015. Yang Yanshan, Ding Qishuo, Ding Weimin, et al. Development of a multi-purpose in-situ experimental platform for tillage tool test[J]. Transactions of the Chinese Society for Agricultural Machinery, 2015. (in Chinese with English abstract)
[31] 孟鳳英,丁啟朔,鹿飛,等. 沖擊作用下粘性土壤破碎體的分形維數(shù)與影響因素[J]. 農(nóng)業(yè)機(jī)械學(xué)報(bào),2009,40(3):108-111. Meng Fengying, Ding Qishuo, Lu Fei, et al. Fragmentation fractal dimensions of cohesive soil under impact and its influencing factors[J]. Transactions of the Chinese Society for Agricultural Machinery, 2009, 40(3): 108-111. (in Chinese with English abstract)
[32] Ding Qishuo, Ding Weimin. Stress wavelets:multi-scale and multi-resolution assessment of soil structure by the drop shatter method[J]. Soil Tillage Research, 2006, 88(1/2): 168-179.
[33] Pei Zhao, Ming-an Shao, Wail Omran,et al. A modified model for estimating the full description of soil particle size distribution[J]. Can. J. Soil Sci., 2013, 93: 65-72.
[34] Diaz-Zorita M, Perfect E, Grove J H. Disruptive methods for assessing soil structure[J]. Soil & Tillage research, 2002,64(s1/2): 3-22.
[35] 趙愛輝,黃明斌,史竹葉. 土壤顆粒分布參數(shù)模型對(duì)黃土性土壤的適應(yīng)性研究[J]. 農(nóng)業(yè)工程學(xué)報(bào),2008,24(1):1-6. Zhao Aihui, Huang Mingbin, Shi Zhuye. Evaluation of parameter models for estimating loess soil particle size distribution[J]. Transactions of the Chinese Society of Agricultural Engineering (Transactions of the CSAE), 2008,24(1): 1-6. (in Chinese with English abstract)
[36] Arvidsson J, Hillerstrom O. Specific draught, soil fragmentation and straw incorporation for different tine and share types[J]. Soil & Tillage research, 2010, 110(1): 154-160.
[37] 丁啟朔,丁為民,潘根興,等. 土壤宏觀力學(xué)結(jié)構(gòu)與精準(zhǔn)耕作[J]. 中國農(nóng)業(yè)科學(xué),2012,45(1):26-33. Ding Qishuo, Ding Weimin, Pan Genxing, et al. Nanjing Agricultural University. Soil macro-structure and precision tillage[J]. China Agriculture Science, 2012, 45(1): 26-33. (in Chinese with English abstract)
[38] 周虎,呂貽忠,李保國. 土壤結(jié)構(gòu)定量化研究進(jìn)展[J]. 土壤學(xué)報(bào),2009,46(3):501-506. Zhou Hu, Lü Yizhong, Li Baoguo. Advancement in the study on quantification of soil structure[J]. Acta Pedologica Sinica,2009, 46(3): 501-506. (in Chinese with English abstract)
[39] 田慎重,王瑜,李娜,等. 耕作方式和秸稈還田對(duì)華北地區(qū)農(nóng)田土壤水穩(wěn)性團(tuán)聚體分布及穩(wěn)定性的影響[J]. 生態(tài)學(xué)報(bào),2013,33(22):7116-7124. Tian Shenzhong, Wang Yu, Li Na, et al. Effects of different tillage and straw systems on soil water-stable aggregate distribution and stability in the North China Plain[J]. Acta Ecologica Sinica, 2013, 33(22): 7116-7124. (in Chinese with English abstract)
[40] 李文昭,周虎,陳效民,等. 基于同步輻射顯微CT研究不同施肥措施下水稻土團(tuán)聚體微結(jié)構(gòu)特征[J]. 土壤學(xué)報(bào),2014,51(1):67-74. Li Wenzhao, Zhou Hu, Chen Xiaomin, et al. Characterization of aggregate microstructures of paddy soil under different patterns of fertilization with synchrotron radiation micro-CT[J]. Acta Pedologica Sinica, 2014, 51(1): 67-74. (in Chinese with English abstract)
[41] 程科,李軍,毛紅玲. 不同輪耕模式對(duì)黃土高原旱作麥田土壤物理性狀的影響[J]. 中國農(nóng)業(yè)科學(xué),2013,46(18):3800-3808. Cheng Ke, Li Jun, Mao Honglin. Effects of different rotational tillage patterns on soil physical properties in rainfed wheat fields of the loess plateau[J]. Scientia Agricultura Sinica, 2013, 46(18): 3800-3808. (in Chinese with English abstract)
[42] 蒲玉琳,林超文,謝德體,等. 植物籬-農(nóng)作坡地土壤團(tuán)聚體組成和穩(wěn)定性特征[J]. 應(yīng)用生態(tài)學(xué)報(bào),2013,24(1):122-128. Pu Yulin, Lin Chaowen, Xie Deti, et al. Composition and stability of soil aggregates in hedgerow-crop slope land[J]. Chinese Journal of Applied Ecology, 2013, 24(1): 122-128. (in Chinese with English abstract)
[43] 祁迎春,王益權(quán),劉軍,等. 不同土地利用方式土壤團(tuán)聚體組成及幾種團(tuán)聚體穩(wěn)定性指標(biāo)的比較[J]. 農(nóng)業(yè)工程學(xué)報(bào),22011,27(1):340-347.Qi Yingchun, Wang Yiquan, Liu Jun, et al. Comparative study on composition of soil aggregates with different land use patterns and several kinds of soil aggregate stability index[J]. Transactions of the Chinese Society of Agricultural Engineering (Transactions of the CSAE), 2011, 27(1): 340-347. (in Chinese with English abstract)
[44] 張斌,許玉芝,李娜,等. 土壤團(tuán)聚結(jié)構(gòu)變化的關(guān)鍵控制過程研究進(jìn)展[J]. 土壤與作物,2014,3(2):41-49. Zhang Bin, Xu Yuzhi, Li Na, et al. Recent development in controlling factors for aggregated soil structure[J]. Soil and Crop, 2014, 3(2): 41-49. (in Chinese with English abstract)
[45] 孔令德,桑正中. 正轉(zhuǎn)旋耕土壤破碎情況的研究[J]. 農(nóng)業(yè)機(jī)械學(xué)報(bào),2001,32(3):31-32. Kong Linde, Sang Zhengzhong. Evaluation of crushing soil from rotary tillage[J]. Transactions of the Chinese Society of Agricultural Machinery, 2001, 32(3): 31-32. (in Chinese with English abstract)
[46] Fernlund J M R. Image analysis method for determining 3-D shape of coarse aggregate[J]. Cement & Concrete Research, 2005, 35(8): 1629-1637.
[47] 車剛,張偉,萬霖,等. 基于滅茬圓盤驅(qū)動(dòng)旋耕刀多功能耕整機(jī)設(shè)計(jì)與試驗(yàn)[J]. 農(nóng)業(yè)工程學(xué)報(bào),2012,28(20):34-40. Che Gang, Zhang Wei, Wan Lin, et al. Design and experiment of multifunctional tillage machine with driven bent blade by stubble ploughing disk[J]. Transactions of the Chinese Society of Agricultural Engineering (Transactions of the CSAE), 2012, 28(20): 34-40. (in Chinese with English abstract)
[48] 李永磊,宋建農(nóng),康小軍,等. 雙輥秸稈還田旋耕機(jī)試驗(yàn)[J].農(nóng)業(yè)機(jī)械學(xué)報(bào),2013,44(6):45-49. Li Yonglei, Song Jiannong, Kang Xiaojun, et al. Experiment on twin-roller cultivator for straw returning[J]. Transactions of the Chinese Society for Agricultural Machinery, 2013,44(6): 45-49. (in Chinese with English abstract)
[49] Bogrekci I, Godwin R J. Development of an image-processing technique for soil tilth sensing[J]. Biosystems Engineering,2007, 97: 323-331.
Digital image processing of mass-size distribution of soil structures in plough layer
Ding Qishuo, Dong Chengsheng, Li Yinian, Qiu Wei, Xue Jinlin, He Ruiyin
(College of Engineering, Nɑnjing Agriculturɑl University, Nɑnjing 210031, Chinɑ)
Abstract:In many instances the basic structural units of plough layer are soil aggregates which are resulted from tillage operation and packed into layers to form the seedbed. Quantification of plough layer is limited to a few basic soil parameters,including cone index, bulk density and porosity. Soil aggregates are assessed with dry sieving. These parameters do not provide the detailed structural information of plough layer. Precision management of plough layer requires that soil structures be quantified with more parameters that are geometrically quantifiable. Quantitative method for tilled-layer soil structures was adopted; the digital images of soil structures were taken, the aggregate mass was calculated and the mass-size distributions of soil structures sampled from a plowed paddy field were studied. The tilt angle of camera was set to 30°, 45°, 60° and 90°,respectively, when taking the photos of soil structures. Soil structures were also measured manually with dry-sieving method for comparison. During the manual measurement of soil aggregates, a caliper was used and both the long and the short axes of the aggregates were measured. The dry-sieving used the nested sieves with the openings of 4, 8, 16, 32, 64 and 128 mm,respectively. Geometrical parameters of soil structures in each size range were calculated with an image-processing program developed in MatLab, including angularity, shape index and rectangle degree. Collected data for the mass-size distribution of the soil aggregates after plowing were also fitted respectively with 3 models, i.e. Weibull model, Rosin-Rammler model and Gaudin-Schuhmann model. It showed that, along with the increase of size range, both angularity and shape index increased,but rectangle degree decreased, meaning that different effects of mechanical operation were induced by different size ranges of soil structures, even though under the same plowing treatment. Detailed analysis on each size range and each tilting angle showed that photos taken with 60otilting angle yielded the best fitting results compared with other tilting angles. The 60otilting angle was the most suitable for camera when used for on-line soil structure monitoring. No significant difference (P>0.05) was observed between digital image processing and manual measurement, proving that the digital image processing was an accurate method to acquire mass-size distributions of soil structure. Compared with Weibull and Rosin-Bammler distribution, Gaudin-Schumann model provided the best fitting between aggregate mass and size, with the R2of 0.98. Digital image processing discriminated soil structures in finer scales and provided a higher precision curve fitting for soil structures compared with dry-sieving method. The variation of the acquired results from dry-sieving was significant due to the large size ranges between adjacent sieve sizes. Unlike the limited methods for tilled-layer soil structure quantification, such as dry sieving, image-processing was capable of not only quantifying the geometrical parameters of soil structures, but also distinguishing and separating soil structures in finer scales, such as 5 mm size range or any other arbitrary scales. This fine scale distinction was helpful in providing more precise modeling on soil structures. The results prove that the image-processing is a powerful tool to calculate geometric parameters of soil structures and discriminate soil structural features in detail.
Keywords:soils; models; image processing; soil tilth; soil structure; digital image processing; mass-size distribution
作者簡介:丁啟朔,男,漢族,江蘇邳州人,教授,博導(dǎo)。南京南京農(nóng)業(yè)大學(xué)工學(xué)院,210031。Email:qsding@njau.edu.cn
基金項(xiàng)目:國家自然科學(xué)基金資助項(xiàng)目(41371238);江蘇優(yōu)勢(shì)學(xué)科建設(shè)資助項(xiàng)目(PAPD)
收稿日期:2015-09-30
修訂日期:2015-12-10
中圖分類號(hào):S152.4
文獻(xiàn)標(biāo)志碼:A
文章編號(hào):1002-6819(2016)-02-0134-07
doi:10.11975/j.issn.1002-6819.2016.02.020