黑土区田块尺度土壤有机质含量遥感反演模型

为了对田块尺度土壤有机质进行空间反演并提高模型精度和稳定性,该文以黑龙江省黑土带41.3 hm~2田块为例,获取2016年5月中下旬两期(受限于拍摄周期和天气原因而选择不同卫星影像,2016年5月17日Landsat 8影像和5月25日Sentinel-2A影像)裸土时期遥感影像和4 m分辨率DEM数据;分析单期影像与土壤有机质(soil organic matter,SOM)的关系,两期影像所包含的土壤含水量变化信息与地形因素对SOM预测模型精度的影响,建立基于BP神经网络的SOM遥感反演模型。结果表明:该田块内SOM含量差异较大;利用单期影像预测SOM时,基于红波段和785~899 nm波段建立的预测模型精度(建模均方根误差RMSE 1.033,检验RMSE 1.079)和稳定性(建模决定系数R2 0.677,检验R20.644)较高;两期影像时,基于红波段和1 570~1 650 nm波段建立的预测模型精度(建模RMSE 0.855,检验RMSE 0.898)和稳定性(建模R2 0.792,检验R2 0.797)显著提高;在两期影像模型基础上,加入地形因子作为输入量,模型精度(建模RMSE 0.492,检验RMSE 0.499)和稳定性(建模R2 0.917,检验R2 0.928)进一步提高。研究成果可为土壤碳库估算和农田精准施肥提供理论与技术支持。     

基于隐马尔科夫模型的深度视频哺乳母猪高危动作识别

哺乳母猪的高危动作和仔猪存活率有密切关系,能直接体现其母性行为能力,而这些高危动作又与其姿态转换的频率、持续时间等密切相关。针对猪舍环境下,环境光线变化、母猪与仔猪黏连、猪体形变等给哺乳母猪姿态转换识别带来的困难。该文以梅花母猪为研究对象,以Kinect2.0采集的深度视频图像为数据源,提出基于Faster R-CNN和隐马尔科夫模型的哺乳母猪姿态转换识别算法,通过FasterR-CNN产生候选区域,并采用维特比算法构建定位管道;利用Otsu分割和形态学处理提取疑似转换片段中母猪躯干部、尾部和身体上下两侧的高度序列,由隐马尔科夫模型识别姿态转换。结果表明,对姿态转换片段识别的精度为93.67%、召回率为87.84%。研究结果可为全天候母猪行为自动识别提供技术参考。     

基于轻量级CNN的植物病害识别方法及移动端应用

为了实现在手机端植物病害叶片检测,对MobileNet和Inception V3 2种轻量级卷积神经网络进行迁移学习,得到2种作物病害分类模型,将2种分类模型分别移植到Android手机端,在识别精度、运算速度和网络尺寸之间进行平衡,选择最优模型。试验表明,MobileNet和InceptionV3在PlantVillage数据集(共38类26种病害)上平均识别率分别是95.02%和95.62%。在自建图像集葡萄病害叶片的识别中MobileNet和InceptionV3平均识别率分别是87.50%、88.06%,Inception V3的整体识别精度略高,但MobileNet在所有类别的识别上均衡性更好;在模型尺寸方面Inception V3的模型尺寸大小为87.5 MB,MobileNet的模型尺寸为17.1 MB,大约是后者5倍;2种模型移植到手机端时,MobileNet和Inception V3的APP所占内存分别是21.5和125 MB;在手机端单张图片的识别时间方面,Inception V3平均计算时间约是174 ms,MobileNet的平均计算时间约是134 ms,后者的平均计算时间比前者快40 ms;在手机端MobileNet相比于Inception V3占用内存更小,运算时间更快。说明MobileNet更适合在手机端进行植物病害识别应用。     

马铃薯薄片干燥过程形态变化三维成像

为研究马铃薯薄片在干燥过程中形态变化规律,该文利用Kinect传感器搭建了图像采集平台,研究其在不同干燥温度下(50、60、70、80℃)的形态变化规律。通过图像采集平台获取马铃薯薄片深度图像和彩色图像,利用彩色图像确定感兴趣区域,对对应区域的深度图像进行灰度值拉伸、阈值分割、边缘去噪处理,进而提取特征,计算出正投影面积的收缩率、深度均值及标准差,以表征马铃薯干燥过程中表面卷曲及平整度等形态指标的变化规律。对不同干燥时间点马铃薯片进行三维图形显示可观察其变化规律明显。统计结果表明:低温(50、60℃)与高温(70、80℃)对马铃薯薄片干燥时的收缩率、卷曲程度具有显著影响(P0.05)。50℃时收缩率为54.97%,80℃时收缩率升高为64.55%;干燥温度与马铃薯片卷曲程度呈先升后降的关系,60℃时卷曲度最大,其深度均值为27.81 mm,80℃时降低到18.86 mm。而四组温度下,马铃薯薄片的平整度具有显著性差异(P0.05),50℃时马铃薯片深度值的标准差为7.99 mm,80℃时降低至5.71mm,说明平整度随着干燥温度升高而增加。该研究可为马铃薯薄片干燥过程中形态变化的检测提供参考,同时为干燥工艺的智能化控制提供技术依据。     

基于汇流累积计算的沟壑密度分析方法

沟壑密度是评价地表侵蚀影响,水土流失情况,进行地貌类型分析等的重要指标。文章基于研究区填洼后的DEM,利用地表汇流累积分析原理和栅格重分类方法,通过研究沟道覆盖区栅格数与汇流累积栅格二值化阈值间的关系,选择合适的阈值用于提取地表沟道。然后结合自然地表高程,分析合理的高程和相对高程栅格重分类阈值并进行高程栅格二值化,将二值化结果作为乘积模板并配合栅格细化算法优化沟道提取结果。最终利用1 km²单位面积规则格网对沟道提取结果进行目标栅格数统计计算,得到地表沟壑密度值。结果显示,该方法基于30 m分辨率DEM提取的西安市地表沟壑密度均值为1.29 km/km²。除雁塔区,其余各行政区统计均值与前人研究结果基本吻合。试验结果证明,基于本方法可以高效较精确地计算该市地表沟壑密度,分析结果可以量化研究区沟壑密度特征值以及通过栅格像元灰度变化体现沟壑密度变化趋势。     

All-around 3D plant modeling system using multiple images and its composition

In this study, we developed an all-around 3D plant modeling system that operates using images and is capable of measuring plants non-destructively without any contact. During the fabrication of this device, we selected a method capable of performing 3D model reconstruction from multiple images. We then developed an improved SfM-MVS (Structure from Motion / Multi-View-Stereo) method that enables 3D reconstruction by simply capturing images with a camera. The resulting image-based method offers a high degree of freedom because the hardware and software can comprise commercially available products, and it permits the use of one or more cameras according to the shape and size of the plant. The advantages of the image-based method are that 3D reconstruction can be conducted at any time as long as the images are already taken, and that the desired locations can be observed, measured, and analyzed from 2D images and a 3D point cloud. The device we developed is capable of 3D measurements and modeling of plants from a few millimeters to 2.4 m of height using this method. This article explains this device, the principles of its composition, and the accuracy of the models obtained from it.     

Improving the efficiency of plant root system phenotyping through digitization and automation

Root system architecture (RSA) determines unevenly distributed water and nutrient availability in soil. Genetic improvement of RSA, therefore, is related to crop production. However, RSA phenotyping has been carried out less frequently than above-ground phenotyping because measuring roots in the soil is difficult and labor intensive. Recent advancements have led to the digitalization of plant measurements; this digital phenotyping has been widely used for measurements of both above-ground and RSA traits. Digital phenotyping for RSA is slower and more difficult than for above-ground traits because the roots are hidden underground. In this review, we summarized recent trends in digital phenotyping for RSA traits. We classified the sample types into three categories: soil block containing roots, section of soil block, and root sample. Examples of the use of digital phenotyping are presented for each category. We also discussed room for improvement in digital phenotyping in each category.     

Image-Based Soil Deformation Measurement in Geotechnical Model Test

Deformation measurement and analysis play an important role in geotechnical model tests. Using the concept of quadrilateral isoperimetric element, the strain filed was achieved from the calculation method based on displacement modes, and the verification results show that this measurement technique has a reliable precision. The soil deformation in model test of square shallow foundation was measured by using this technique, and the results show that the measurement system based on digital image correlation method can realize the whole process quantitative measurement of soil displacement field and shear strain filed. This deformation measurement technique provides a tool for the qualitative and quantitative analysis of soil micromechanics characteristics.     

“麦德姆”造成山东暴雨的卫星水汽图像分析

为提高台风暴雨的预报准确率,降低对农业生产的影响,本研究利用常规观测资料和FY-2E卫星云图,分析台风"麦德姆"造成山东暴雨的成因。结果表明:前期强降水是由台风倒槽造成,后期中、低纬度系统相互作用导致半岛地区出现大暴雨;水汽图像上的亮区对应对流层中上部湿区,亮区与低层水汽通量散度大值区和500 h Pa总温度场高能舌重合,对台风路径的预报和暴雨落区有较好的指示意义;水汽图像上系统的形态和灰度变化,能更有效地监测中、低纬度系统相互影响导致的过程,叠加云导风产品能更清晰地监测台风进入中纬度变性为温带气旋的风场与云系变化,云导风产品在台风预报分析中能发挥重要作用。     

基于差分投影法实现仓储害虫图像的精确定位

针对仓储害虫图像的特点,提出了一种基于投影法的图像目标快速定位算法.本算法的优点是特征提取简单,计算量小,速度快;在处理时间上可以满足应用的实时性要求.通过在对仓储害虫差分图像处理中的试验证明,该算法能够快速地识别出实时采集的图像中的目标位置,定位精确.