期刊文献+
共找到20篇文章
< 1 >
每页显示 20 50 100
加权高斯减法拟合的亮度可控土壤图像增强 被引量:3
1
作者 曾绍华 赵秉渝 +2 位作者 王帅 陈亚楠 朱德利 《光子学报》 EI CAS CSCD 北大核心 2022年第4期348-367,共20页
为了将野外不同光照环境机器视觉采集的土壤图迁移到特定亮度,消除成像条件不一致对紫色土土种识别的影响,提出土壤图像亮度可控增强算法。其方法是构建左局部区域的高斯拟合优化模型对土壤图像Y分量的直方图进行拟合,求拟合剩余部分,... 为了将野外不同光照环境机器视觉采集的土壤图迁移到特定亮度,消除成像条件不一致对紫色土土种识别的影响,提出土壤图像亮度可控增强算法。其方法是构建左局部区域的高斯拟合优化模型对土壤图像Y分量的直方图进行拟合,求拟合剩余部分,继续对剩余部分的左局部区域高斯拟合,直到剩余部分足够小,获得多次高斯拟合的加权拟合曲线;然后在加权高斯减法拟合曲线中引入目标亮度实现图像亮度迁移,完成土壤图像亮度可控增强;最后根据色比不变性原理,对U、V分量颜色校正,获得土壤增强图像。实验结果显示:以原图亮度为基点,正负30个亮度灰度级为所提算法的有效增强变换范围;所提算法比目前文献可查到的仅有的2个可控亮度图像增强对比算法的土壤图像增强变换的精度高,失真度小。 展开更多
关键词 图像增强 亮度可控 加权高斯减法拟合 土壤图像 机器视觉
下载PDF
基于减法直方图算法的土壤图像阴影检测 被引量:1
2
作者 曾绍华 王琪 +2 位作者 佘春燕 王帅 罗达璐 《西南大学学报(自然科学版)》 CAS CSCD 北大核心 2021年第10期167-180,共14页
自然环境下机器视觉采集的土壤图像存在阴影,土壤图像阴影检测将消除或者减弱阴影对后一步的子图分割及土种识别的影响.为了提高土壤图像阴影和非阴影的区分度,实现阴影检测,重构了比率(α)特征,并通过高斯平滑分别获得亮度(I)和比率(α... 自然环境下机器视觉采集的土壤图像存在阴影,土壤图像阴影检测将消除或者减弱阴影对后一步的子图分割及土种识别的影响.为了提高土壤图像阴影和非阴影的区分度,实现阴影检测,重构了比率(α)特征,并通过高斯平滑分别获得亮度(I)和比率(α)特征直方图的2个主峰值点,缩小分割阈值搜索区间以减少减法直方图求高保留率点(F)的次数;再引入拉伸因子对2个特征的保留直方图拉伸,增大阴影和非阴影保留率差异,以获取其高保留率点F;最后,在搜索区间构建亮度(I)和比率(α)特征的64等份中位点及其对应F的2条折线的交点,即亮度(I)和比率(α)特征的F值相等点,获得交点对应的亮度(I)和比率(α)特征值为阴影检测阈值,分割土壤图像阴影和非阴影.仿真实验显示:本文算法相较于对比实验具有更好的分割精度,能自适应检测土壤图像阴影,算法是有效的. 展开更多
关键词 阴影检测 减法直方图 土壤图像 机器视觉
下载PDF
基于Sigmoid曲线拟合的亮度可控土壤图像增强
3
作者 徐昌莉 曾绍华 +3 位作者 李娇 刘国一 詹雪萍 龙伍 《重庆师范大学学报(自然科学版)》 CAS 北大核心 2024年第1期86-99,共14页
图像在相同条件下表征土壤特征会提高土壤图像识别土种的精度。对自然环境下机器视觉采集的土壤图像亮度可控增强,将不同光照条件采集的土壤图像转换为近似于该土壤在某些特定光照条件下采集的具有一定亮度的真实土壤图像,能消除或减弱... 图像在相同条件下表征土壤特征会提高土壤图像识别土种的精度。对自然环境下机器视觉采集的土壤图像亮度可控增强,将不同光照条件采集的土壤图像转换为近似于该土壤在某些特定光照条件下采集的具有一定亮度的真实土壤图像,能消除或减弱光照对后续土壤图像土种识别的影响。因此,应用Sigmoid曲线对土壤图像亮度(Y)分量的累积概率密度(cumulative distribution function,cdf)曲线拟合;然后,构建目标亮度逼近优化模型,迁移拟合的Sigmoid曲线逼近目标亮度;再依据像素的邻域信息对相同亮度的像素排序、迁移,实现土壤图像的亮度可控增强;最后,利用高斯卷积核提取色调(U)、饱和度(V)分量的低频分量,并基于色比不变性原理与原始土壤图像的邻域信息对增强土壤图像的U、V分量高低频分别增强,完成颜色校正,并融合增强亮度分量,获得增强的彩色土壤图像。实验结果表明,提出算法对完全重合的亮度不同成对真实土壤图像做有目标增强实验,增强后的土壤图像与真实目标土壤图像对应像素Y、U、V分量差的标准差均值分别为14.313 7、1.323 2、2.110 5,峰值信噪比均值为29.820 9;与对比算法2-D HS、WGSF比较,计算对比算法增强后的土壤图像与真实目标土壤图像对应像素Y、U、V分量差的标准差均值,提出算法比对比算法分别降低了0.767 7~4.762 9、0.052 4~1.110 4、0.071 4~1.272 0。所提算法对土壤图像亮度可控增强的精度高,失真度小,有效亮度增强范围为[-35,35],实验证明算法是有效的。 展开更多
关键词 土壤图像增强 亮度可控 Sigmoid曲线拟合 颜色校正
原文传递
非对称广义高斯亮度迁移与局部信息融合的土壤图像增强
4
作者 邓亿 曾绍华 +2 位作者 詹林庆 陈亚楠 刘国一 《重庆师范大学学报(自然科学版)》 CAS 北大核心 2023年第3期139-154,共16页
【目的】为减小土壤图像成像环境条件差异对机器视觉识别土种精度的影响,提出对野外自然环境下采集的土壤图像可控光照增强。【方法】首先,用非对称广义高斯曲线拟合土壤图像V分量直方图,并在拟合曲线中引入目标迁移量,完成图像亮度迁移... 【目的】为减小土壤图像成像环境条件差异对机器视觉识别土种精度的影响,提出对野外自然环境下采集的土壤图像可控光照增强。【方法】首先,用非对称广义高斯曲线拟合土壤图像V分量直方图,并在拟合曲线中引入目标迁移量,完成图像亮度迁移,实现土壤图像全局亮度可控增强;然后,利用全局和局部信息估计图像在局部区域上的光照权重,引入目标亮度,根据权重确定局部增量,并将局部增量叠加到原始V分量,实现基于局部增量的土壤图像亮度增强;再利用sigmoid曲线,将非对称广义高斯曲线亮度迁移获得全局亮度增强结果与基于局部增量的亮度增强结果融合,获得土壤图像的亮度可控增强。最后,根据色比不变性原理,分别对原始土壤图像R、G、B分量进行颜色校正。【结果】仿真实验表明:土壤图像从低亮度向高亮度迁移时,增强图像与目标图像在V分量上各对应像素的亮度差异均值为10.5267,与目标图像亮度均值差异为0.2451;从高亮度向低亮度迁移时,各对应像素的亮度差异均值为10.7430,与目标图像亮度均值差异为0.2721;本文所提算法在亮度控制上较其他算法具有更高的精度;主观质量评价表明,以原图亮度为基准,[-30,30]为土壤图像有效增强范围;【结论】所提方法能够实现亮度可控的土壤图像增强,算法是有效的。 展开更多
关键词 图像增强 亮度控制 土壤图像 直方图规定化 机器视觉
原文传递
基于无人机遥感测绘技术的土壤有机污染监测方法研究
5
作者 冯健 《环境科学与管理》 CAS 2024年第4期128-132,共5页
为实时监测土壤有机污染,提出基于无人机遥感测绘技术的土壤有机污染监测方法。设计无人机遥感测绘装置,获取监测区域土壤光谱数据,应用小波变换算法去除土壤光谱数据的噪声信息,采用惩罚最小二乘法去除土壤光谱本底信息,通过LAR算法选... 为实时监测土壤有机污染,提出基于无人机遥感测绘技术的土壤有机污染监测方法。设计无人机遥感测绘装置,获取监测区域土壤光谱数据,应用小波变换算法去除土壤光谱数据的噪声信息,采用惩罚最小二乘法去除土壤光谱本底信息,通过LAR算法选择与提取光谱数据特征变量,衡量其与已知有机污染物质光谱特征变量的相似程度,判定监测区域土壤是否存在有机污染,并确定有机污染物质种类。实验数据显示:应用提出方法获得的监测区域土壤有机污染判定结果与实际结果保持一致,土壤有机污染物质监测因子最大值为0.98。 展开更多
关键词 土壤遥感图像 污染监测 特征提取 无人机遥感测绘技术 有机污染
下载PDF
遥感图像植被判读与土壤判读 被引量:3
6
作者 刘王君 贾明 《科技情报开发与经济》 2005年第6期141-143,共3页
通过遥感图像判断识别各种目标,是达到遥感应用目的的一个重要环节,而判读植被图像和土壤图像对监测生态环境变化具有十分重要的意义。介绍了主要植被类型的影像特征和遥感图像植被判读的方法,以及航空像片的土壤判读和卫星图像的土壤... 通过遥感图像判断识别各种目标,是达到遥感应用目的的一个重要环节,而判读植被图像和土壤图像对监测生态环境变化具有十分重要的意义。介绍了主要植被类型的影像特征和遥感图像植被判读的方法,以及航空像片的土壤判读和卫星图像的土壤判读方法。 展开更多
关键词 遥感图像 图像判读 植被图像 土壤图像 生态环境监测
下载PDF
基于高速摄像下圆锥螺旋挖藤清土土壤运动图像分析
7
作者 刘松 郭辉 +3 位作者 刘书艳 赵占军 张豪 杨宛章 《中国农机化学报》 2015年第5期292-295,共4页
对于我国北方地区葡萄种植情况来说,葡萄果园入春时期挖藤清土是葡萄生产种植过程中一项重要的农艺要求。本文借助高速摄像技术及图像处理技术对圆锥螺旋挖藤清土装置土壤运动规律进行研究,分析了土壤颗粒特征目标运动,为进一步研究设... 对于我国北方地区葡萄种植情况来说,葡萄果园入春时期挖藤清土是葡萄生产种植过程中一项重要的农艺要求。本文借助高速摄像技术及图像处理技术对圆锥螺旋挖藤清土装置土壤运动规律进行研究,分析了土壤颗粒特征目标运动,为进一步研究设计圆锥螺旋式入春葡萄挖藤机提供实验依据。 展开更多
关键词 葡萄果园 入春时期挖藤清土 高速摄像 土壤运动图像分析
下载PDF
土壤显微灰度图像三维重建的研究
8
作者 鲍苏苏 李强 李吉桂 《计算机科学》 CSCD 北大核心 2003年第7期177-179,共3页
It is very important to analyze the microstructure of soil for keep away the groundsill fall down and distor-tion. After the soil samples are dried in the vacancy container, the toil degree image is taken under the el... It is very important to analyze the microstructure of soil for keep away the groundsill fall down and distor-tion. After the soil samples are dried in the vacancy container, the toil degree image is taken under the electron micro-scope. As usual, the soil micro-image is processed with 2 D method. In that way, many characters of soil micro-im-age are loosed. In this paper we introduce how to realize the 3D reconstruction of soil micro-image using OpenGL. Bythis way, the 3D structure of image can be sighted directly. More stability gist can be found for judging that ground-sill has being reached the criterion of sinking whether or not. 展开更多
关键词 图像处理 图像像素灰度值 三维重建 土壤显微灰度图像 OPENGL 三维场景
下载PDF
土壤显微灰度图像三维重建的研究
9
作者 李强 鲍苏苏 《安阳师范学院学报》 2003年第2期40-41,49,共3页
在饱和粘土上建造公路,工程界最关心的是地基稳定和变形问题。分析土壤微观结构对防范地基的坍塌、下陷和变形有非常重要的意义。通过对取到的土样在真空下进行干燥,用Hitachis_3500N电子扫描电镜进行拍摄得到了一些土壤灰度图像。通... 在饱和粘土上建造公路,工程界最关心的是地基稳定和变形问题。分析土壤微观结构对防范地基的坍塌、下陷和变形有非常重要的意义。通过对取到的土样在真空下进行干燥,用Hitachis_3500N电子扫描电镜进行拍摄得到了一些土壤灰度图像。通常采用的方法都是对电子显微镜拍摄的照片进行二维处理,既丢失了不少特征,又不直观。本文主要介绍如何利用OpenGL来实现土壤微观结构图像的三维重建,从而获得对土壤微观结构特征进行分析的一种新的分析方法。 展开更多
关键词 公路工程 土壤灰度图像 三维重建 微观结构 OPENGL 地基稳定 地基变形
下载PDF
基于CT扫描技术的土壤孔隙定量表达优化 被引量:20
10
作者 赵玥 韩巧玲 赵燕东 《农业机械学报》 EI CAS CSCD 北大核心 2017年第10期252-259,共8页
现有土壤孔隙量化方法主要通过图像处理软件实现孔隙结构的辨识与分析,此类通用的图像处理软件或医学处理软件未考虑土壤内部物质的复杂多变性以及孔隙结构的细小和不规则性,从而导致孔隙分割精度低进而量化误差大,为解决这一问题,本文... 现有土壤孔隙量化方法主要通过图像处理软件实现孔隙结构的辨识与分析,此类通用的图像处理软件或医学处理软件未考虑土壤内部物质的复杂多变性以及孔隙结构的细小和不规则性,从而导致孔隙分割精度低进而量化误差大,为解决这一问题,本文针对土壤CT图像的特点提出了一种孔隙量化方法。该方法主要包括图像处理和量化分析两部分:选用自适应中值滤波算法去除噪声对孔隙边缘的影响,并采用迭代最佳阈值法与Canny边缘检测算子相结合的方法,准确识别出土壤孔隙结构及轮廓线;运用数学统计方法定量研究土壤孔隙率、孔隙数目、分形维数、成圆率等几何指标,用以揭示孔隙结构的复杂性和不规则性,实现对土壤孔隙的量化分析。最后,以冻融循环作用下的土壤为应用对象验证该方法性能。结果表明,本文方法能精确地定位孔隙轮廓,有效地分割孔隙结构,而且通过多种孔隙几何指标的量化可揭示出冻融循环作用对土壤结构的影响,为孔隙几何特征和空间特征的量化表达奠定了基础。 展开更多
关键词 土壤断层扫描图像 孔隙结构 图像处理技术 孔隙量化
下载PDF
基于冻融循环的土壤物理状态的自动判别 被引量:1
11
作者 韩巧玲 赵玥 姚立红 《浙江农业学报》 CSCD 北大核心 2017年第7期1189-1194,共6页
以东北典型黑土区土壤为研究对象,采用CT扫描技术与图像处理相结合的方法,通过灰度共生矩阵和主成分分析法提取图像特征,计算测试图像特征向量与训练图像特征向量间的欧氏距离,以此为依据,实现对经历不同冻融循环次数土壤的自动判别。... 以东北典型黑土区土壤为研究对象,采用CT扫描技术与图像处理相结合的方法,通过灰度共生矩阵和主成分分析法提取图像特征,计算测试图像特征向量与训练图像特征向量间的欧氏距离,以此为依据,实现对经历不同冻融循环次数土壤的自动判别。研究结果表明:面向土壤CT图像数据库,基于灰度共生矩阵和主成分分析提取的图像特征,均能实现对土壤的自动判别,但灰度共生矩阵法的判别正确率要高于主成分分析法。 展开更多
关键词 土壤CT图像 灰度共生矩阵 主成分分析 欧氏距离 判别正确率
下载PDF
Historical Development in Soil Micromorphological Imaging 被引量:2
12
作者 A.R.Mermut 《Journal of Mountain Science》 SCIE CSCD 2009年第2期107-112,共6页
The book "micropedolog" by Kubieana and a large number of publications has induced many people to practice soil micromorphology. Quantification of the soil fabric and its components was a major challenge. The use of... The book "micropedolog" by Kubieana and a large number of publications has induced many people to practice soil micromorphology. Quantification of the soil fabric and its components was a major challenge. The use of the image analyses in soil science was a breakthrough. Attempts to make soil thin sections go back to the beginning of the 2oth century. Microscopic techniques and recently high resolution electron microscope and use of computer assisted imaging techniques enabled the in vitro study of soils in three dimensional levels. It is now possible to store and process massive amounts of data. Micro- morphological concepts and techniques are applied in paleopedological, ecological, and archaeological studies. The aim of this work was to examine soil micromorphological imaging in historical perspective. 展开更多
关键词 Soil micromorphology quantification imaging techniques HISTORY
下载PDF
Soil Organic Carbon Stock as Affected by Land Use/Cover Changes in the Humid Region of Northern Iran 被引量:3
13
作者 Samereh FALAHATKAR Seyed Mohsen HOSSEINI +2 位作者 Abdolrassoul SALMAN MAHINY Shamsollah AYOUBI WANG Shao-qiang 《Journal of Mountain Science》 SCIE CSCD 2014年第2期507-518,共12页
【Title】【Author】This study was conducted to determine the changes in the soil carbon stocks as influenced by land use in a humid zone of Deylaman district (10,876 ha), a mountainous region of northern Iran. For t... 【Title】【Author】This study was conducted to determine the changes in the soil carbon stocks as influenced by land use in a humid zone of Deylaman district (10,876 ha), a mountainous region of northern Iran. For this, land use maps were produced from TM and ETM+ images for 1985, 2000 and 2010 years; and this was supplemented by field measurement of soil carbon in 2010. The results showed that the mean soil organic carbon (SOC) density was 6.7±1.8 kg C m-2, 5.2±3.4 kg C m-2 and 3.2±1.8 kg C m-2 for 0-20 cm soil layer and 4.8±1.9 kg C m-2, 3.1±2 kg C m-2 and 2.7±1.8 kg C m-2 for 20-40 cm soil layer in forest, rangeland and cultivated land, respectively. During the past 25 years, 14.4% of the forest area had been converted to rangeland; and 28.4% of rangelands had been converted to cultivated land. According to the historical land use changes in the study area, the highest loss of SOC stocks resulted from the conversion of the forest to rangeland (0.45×104 Mg C in 0-40 cm depth layer); and the conversion of rangeland to cultivated land (0.37×104 Mg C in 0-40 cm), which typically led to the loss of soil carbon in the area studied. The knowledge on the historical land use changes and its influence on overall SOC stocks could be helpful for making management decision for farmers and policy managers in the future, for enhancing the potential of C sequestration in northern Iran. 展开更多
关键词 Soil organic carbon stocks Land cover Land use Iran
下载PDF
Land Parcel Land Use History as a Key to Site Selectionfor Documenting Soil Contamination Risk: a Case Study from Australian Suburbia
14
作者 YU Jie Ursula Pietrzak Jim Peterson 《Geo-Spatial Information Science》 2005年第4期257-261,275,共6页
In that orcharding in early to mid twentieth century southeastern Australia involved use of certain heavy metal and As compounds in regular pest control spray procedures, some interest attaches to the possibility that... In that orcharding in early to mid twentieth century southeastern Australia involved use of certain heavy metal and As compounds in regular pest control spray procedures, some interest attaches to the possibility that these landparcels are underlain by soils with above background Cu, Pb and As levels. Interpretation of Land cover changes allowed land parcels previously occupied by orchards to be identified in the 1950s through time series air photos. A comparison of soil analysis results referring to soil samples from control sites, and from land parcels formerly occupied by orchardists, shows that contamination (above background) levels of cations in the pesticides can be found in the top 6 cm of former orchard soils. It is clear that digital spatial data handling and culturally informed air photo interpretation has a place in soil contamination studies, land use planning (with particular reference to re development) and in administration of public health. 展开更多
关键词 aerial photography digital orthophoto image soil contamination soil analysis land-use planning public health
下载PDF
Identifying soil structure along headwater hillslopes using ground penetrating radar based technique 被引量:7
15
作者 HAN Xiao-le LIU Jin-tao +1 位作者 ZHANG Jun ZHANG Zhi-cai 《Journal of Mountain Science》 SCIE CSCD 2016年第3期405-415,共11页
Soil structure plays an important role in understanding soil attributes as well as hydrological processes. Effective method to obtain high quality soil map is therefore important for both soil science research and soi... Soil structure plays an important role in understanding soil attributes as well as hydrological processes. Effective method to obtain high quality soil map is therefore important for both soil science research and soil work ability improvement. However,traditional method such as digging soil pits is destructive and time-consuming. In this study, the structure of headwater hillslopes from Hemuqiao catchment(Taihu Basin, China) have been analyzed both by indirect(ground penetrating radar, GPR) and direct(excavation or soil auger) methods. Four transects at different locations of hillslopes in the catchment were selected for GPR survey. Three of them(#1, #2, and #3) were excavated to obtain fullscale soil information for interpreting radar images.We found that the most distinct boundary that can be detected by GPR is the boundary between soil and underlain bedrock. In some cases(e.g., 8-17 m in transect #2), in which the in situ soil was scarcely affected by colluvial process, different soil layers can be identified. This identification process utilized the sensitive of GPR to capture abrupt changes of soil characteristics in layer boundaries, e.g., surface organic layer(layer #1) and bamboo roots layer(layer#2, contain stone fragments), illuvial deposits layer(layer #3) and regolith layer(layer #4). However, in areas where stone fragments were irregularly distributed in the soil profile(highly affected bycolluvial and/or fluvial process), it was possible to distinguish which part contains more stone fragments in soil profile on the basis of reflection density(transect #3). Transect #4(unexcavated) was used to justify the GPR method for soil survey based on experiences from former transects. After that, O horizon thickness was compared by a hand auger.This work has demonstrated that GPR images can be of a potential data source for hydrological predictions. 展开更多
关键词 Ground Penetrating Radar(GPR) Hillslope Soil structure
下载PDF
Earth Observation as a Support to Improve Water Use in Irrigated Agriculture 被引量:2
16
作者 E. V. Palacios L. S. Palacios +1 位作者 J. C. Rodriguez J. E. S. Palacios 《Journal of Agricultural Science and Technology》 2010年第6期1-9,共9页
By the end of sixties, several satellites were launched to observe the earth; among them the Landsat missions have been important to study natural resources, especially in agriculture. Currently, Landsat 7, with some ... By the end of sixties, several satellites were launched to observe the earth; among them the Landsat missions have been important to study natural resources, especially in agriculture. Currently, Landsat 7, with some problems, and Landsat 5 are useful to evaluate crop development. A project called "Participatory multi-Level EO-assisted tools for Irrigation water management and Agricultural Decision-Support" (PLEIADES) was financed by the European Commission. This project considers eleven countries and 23 research institutions, Mexico being one of the selected countries. Landsat images are a basic tool of this project. Some of the results obtained in this study are presented in this article. By using w^getation indices obtained by the combinations of reflectances in different bands of the electromagnetic spectrum, it has been possible to estimate the behavior of several biophysical parameters of crops which are used to evaluate plant development, water stress, spatial soil variation, and the effect of plagues and crop diseases. To facilitate this process, a visor was developed, named SPIDER (System of Participatory Information Decision support and Expert knowledge for River basin management) whose use in this work is explained. 展开更多
关键词 SATELLITES reflectances vegetation indices water stress SPIDER.
下载PDF
Local Data Analysis for Eliminating End Restraint of Triaxial Specimen 被引量:1
17
作者 刘潇 邵龙潭 郭晓霞 《Transactions of Tianjin University》 EI CAS 2013年第5期372-380,共9页
A data processing method was proposed for eliminating the end restraint in triaxial tests of soil. A digital image processing method was used to calculate the local deformations and local stresses for any region on th... A data processing method was proposed for eliminating the end restraint in triaxial tests of soil. A digital image processing method was used to calculate the local deformations and local stresses for any region on the surface of triaxial soil specimens. The principle and implementation of this digital image processing method were introduced as well as the calculation method for local mechanical properties of soil specimens. Comparisons were made between the test results calculated by the data from both the entire specimen and local regions, and it was found that the deformations were more uniform in the middle region compared with the entire specimen. In order to quantify the nonuniform characteristic of deformation, the non-uniformity coefficients of strain were defined and calculated. Traditional and end-lubricated triaxial tests were conducted under the same condition to investigate the effects of using local region data for deformation calculation on eliminating the end restraint of specimens. After the statistical analysis of all test results, it was concluded that for the tested soil specimen with the size of 39.1 mm × 80 ram, the utilization of the middle 35 mm region of traditional specimens in data processing had a better effect on eliminating end restraint compared with end lubrication. Furthermore, the local data analysis in this paper was validated through the comparisons with the test results from other researchers. 展开更多
关键词 end restraint triaxial test digital image processing end lubrication local data analysis
下载PDF
基于子图选择的土种识别——以重庆市璧山区紫色土为例
18
作者 陈怡达 曾绍华 +3 位作者 吴雪 王帅 刘国一 周鹏 《重庆师范大学学报(自然科学版)》 CAS 北大核心 2024年第3期59-72,共14页
针对机器视觉野外自然条件下采集土壤图像包含阴影、空洞、缝隙等可能对土种识别产生影响的因素,和土壤图像采集、标记成本高,样本量小的问题,提出一种基于土壤子图选择的土种识别方法。该方法以局部阴影最小化及空间距离最大化为准则,... 针对机器视觉野外自然条件下采集土壤图像包含阴影、空洞、缝隙等可能对土种识别产生影响的因素,和土壤图像采集、标记成本高,样本量小的问题,提出一种基于土壤子图选择的土种识别方法。该方法以局部阴影最小化及空间距离最大化为准则,构造最优化模型;并基于最大最小距离算法思想,迭代更新距离矩阵求解优化模型,获得选择子图中心点,构建土壤子图数据集。实验结果显示:土壤子图选择算法构建的数据集在3个不同深度ResNet模型下训练、识别测试,均有较好的训练、识别精度;在土壤子图选择α参数为1、子图尺寸为224时,在浅层网络ResNet-18下能达到最佳测试结果,训练的网络模型在验证集识别准确率为92.48%,测试集识别准确率为92.95%,相较于土壤原图数据集的最佳土种识别结果提升46.65%;利用最小外接矩阵能加速相较于不使用加速,子图选择算法提升运算速度38.88%。基于子图选择的土种识别大幅度提升了土种识别准确率,证明算法是有效的。 展开更多
关键词 土种识别 子图选择 深度学习 土壤图像
原文传递
Studying the Condition of Soil Protection Agrolandscape in Ukraine Using Remote Sensing Methods
19
作者 Stanislav Truskavetsky Tetiana Byndych Alexandr Sherstyuk Kostiantyn Viatkin 《Journal of Agricultural Science and Technology(A)》 2015年第4期235-240,共6页
The article reviews the scientific approaches to monitoring of soil condition on the soil protection agrolandscape. In 1980s, the contour-meliorative soil protection system was established on the selected fields in Uk... The article reviews the scientific approaches to monitoring of soil condition on the soil protection agrolandscape. In 1980s, the contour-meliorative soil protection system was established on the selected fields in Ukraine. The objective of the current research was to determine the capabilities of satellite survey to identify the changes of soil cover that had occurred on these fields during the past 25 years. Soil erosion processes are very dynamic, therefore it is essential to use time-series of operative satellite images to track those changes. Rills on the fields, caused by water erosion, are clearly identified on high-resolution satellite data. Erosion causes the decrease of humus content, which affects soil reflection values. This in turn leads to a corresponding change of color shade on satellite images. The research allowed to determine correlation between remote sensing data and soil organic carbon content and to acquire a mathematical model which describes this correlation. The condition of the agrolandscape soils was assessed using the regression model, which helped to evaluate erosion risk for different areas of the test polygon. The visual interpretation of satellite imagery led to a conclusion about a damaging effect of erosion on protective forest belts and accordingly on fields' soil cover and crops. Visual analysis results were approved by field research. Photos taken during the field research indicate an unsatisfactory status of forest belts and a devastating effect of eroding water flows. These are the results of irresponsible land use and constant violation of methodical principles of the contour-meliorative system organization. The article concludes that the use of time-series of high-resolution satellite imagery allows monitoring the condition of soil protection agrolandscape, in particular the forest belts' status soil cover conditions and their change over time. The research results can be used as an informational basis for the soil protection agrolandscape monitoring system. 展开更多
关键词 Soil cover space imagery remote sensing anti-erosion agrolandscape soil organic matter MONITORING modeling.
下载PDF
Soil Pore Characterization Using Free Software and a Portable Optical Microscope 被引量:1
20
作者 L. F. PIRES F. S. BORGES +1 位作者 S. PASSONI A. B. PEREIRA 《Pedosphere》 SCIE CAS CSCD 2013年第4期503-510,共8页
Total porosity (TP), determined by image analysis, pore type and pore size distribution were evaluated on impregnated soil blocks from an undisturbed Brazilian sandy loam soil using a digital portable optical micros... Total porosity (TP), determined by image analysis, pore type and pore size distribution were evaluated on impregnated soil blocks from an undisturbed Brazilian sandy loam soil using a digital portable optical microscope. The free software Image J (version 1.40g) was used for image analysis. Procedures for soil image collection and analysis were presented. The image analysis allowed the evaluation of pore sizes with diameters ranging from 20 to 〉 1 000 μm. The following types of pores were also obtained: rounded, elongated and intermediate. The results M[owed the characterization of the soil as moderately porous (TP = 21.6%). Rounded, intermediate and elongated pores were responsible for 11.6%, 31.7% and 56.7% of TP. In relation to pore size 51.1% of TP was in the 100-500μm size class and a third of TP came from the pores larger than 500μm. 展开更多
关键词 image analysis soil porosity soil structure pore size distribution
原文传递
上一页 1 下一页 到第
使用帮助 返回顶部