Crop area monitoring within an advanced agricultural information system

This paper describes a framework for an image processing procedure for operational agricultural crop area estimation. This operational framework has been conceived within the development of an Advanced Agricultural Information System (AAIS) for the “Regione del Veneto “ (RdV ‐ Veneto Region) in northeastern Italy. The objective of this program is to develop the ability to generating timely and accurate area estimates and production information for four major agricultural crops: soybeans, sugar beets, corn, and small grains. AAIS uses state of the art methods in remote sensing and geographic information systems (GIS) technology and integrates a variety of data types including satellite imagery. This paper describes the methodology developed for image and ancillary data processing for the production of crop area statistics. Using a combination of standard unsupervised classification and GIS operations that incorporate knowledge about the agricultural system, a “sequential masking” classification procedure was derived. This sequential masking procedure yielded crop classification accuracies that at the study site level range between 76% and 99% depending on the crop under study. We believe that classification accuracies will improve with full system implementation, along with the incorporation of new and/or improved thematic information and operational experience using AAIS‐based estimation.     

面向对象与卷积神经网络模型的GF-6 WFV影像作物分类

GF-6 WFV影像是中国首颗带有红边波段的中高分辨率8波段多光谱卫星的遥感影像,对于其影像及红边波段对作物分类影响的研究利用亟待展开。本文结合面向对象和深度学习提出一种适用于GF-6 WFV红边波段的卷积神经网络(RE-CNN)遥感影像作物分类方法。首先采用多尺度分割和ESP工具选择最佳分割参数完成影像分割,通过面向对象的CART决策树消除椒盐现象的同时提取植被区域,并转化为卷积神经网络的输入数据,最后基于Python和Numpy库构建的卷积神经网络模型(RE-CNN)用于影像作物分类及精度验证。有无红边波段的两组分类实验结果表明:在红边波段组,卷积神经网络(RE-CNN)作物分类识别取得了较好的效果,总体精度高达94.38%,相比无红边波段组分类精度提高了2.83%,验证了GF-6 WFV红边波段对作物分类的有效性。为GF-6 WFV红边波段影像用于作物的分类研究提供技术参考和借鉴价值。     

基于高分二号卫星数据的农作物分类方法研究

针对目前高分二号卫星数据(GF-2)有较高的空间分辨率而在农业领域应用较少和农作物分类普遍存在"同谱异物"和"同物异谱"的现象,以辽宁省沈阳市苏家屯区以西的新开河村周边为试验基地,利用最佳波段组合指数法(OIF)对所选取的高分二号(GF-2)卫星数据的纹理特征和植被指数以及波段信息进行筛选,选取最佳的波段组合,以增加分类信息、减少数据冗余。最后,针对筛选后的数据,使用最大似然法进行分类,得到农作物的分类结果。结果表明,利用该方法对农作物进行分类,分类精度得到了一定程度的提高,为目前大规模农作物种植面积的精确、迅速统计提供了一套可行的方案。     

Evaluation of Landsat TM data for forest cover type and landuse classification in subtropical forests of Kumaon Himalaya (U.P.)

LANDSAT-TM has been evaluated for forest cover type and landuse classification in subtropical forests of Kumaon Himalaya (U.P.) Comparative evaluation of false colour composite generated by using various band combinations has been made. Digital image processing of Landsat-TM data on VIPS-32 RRSSC computer system has been carried out to stratify vegetation types. Conventional band combination in false colour composite is Bands 2, 3 and 4 in Red/Green/Blue sequence of Landsat TM for landuse classification. The present study however suggests that false colour combination using Landsat TM bands viz., 4, 5 and 3 in Red/Green/Blue sequence is the most suitable for visual interpretation of various forest cover types and landuse classes. It is felt that to extract full information from increased spatial and spectral resolution of Landsat TM, it is necessary to process the data digitally to classify land cover features like vegetation. Supervised classification using maximum likelihood algorithm has been attemped to stratify the forest vegetation. Only four bands are sufficient enough to classify vegetaton types. These bands are 2,3,4 and 5. The classification results were smoothed digitaly to increase the readiability of the map. Finally, the classification carred out using digital technique were evaluated using systematic sampling design. It is observed that forest cover type mapping can be achieved upto 80% overall mapping accuracy. Monospecies stand Chirpine can be mapped in two density classes viz., dense pine (<40%) with more than 90% accuracy. Poor accuracy (66%) was observed while mapping pine medium dense areas. The digital smoothening reduced the overall mapping accuracy. Conclusively, Landsat-TM can be used as operatonal sensor for forest cover type mapping even in complex landuse-terrain of Kumaon Himalaya (U.P.)     

An experimental comparison between KELM and CART for crop classification using Landsat-8 OLI data

The operational land imager (OLI) is the latest instrument in the Landsat series of satellite imagery, which officially began normal operations on 30 May 2013. The OLI includes two bands that are not on the thematic mapper series of sensors aboard Landsat-5 and 7; a cirrus band and a coastal/aerosol band. This paper compares the classification and regression tree and the kernel-based extreme learning machine (KELM) for mapping crops in Hokkaido, Japan, using OLI data, except the cirrus band and the pan band. The OLI data acquired on 8 July 2013 was used for crop classification of beans, beets, grassland, maize, potatoes and winter wheat. The KELM algorithm performed better in this study and achieved overall accuracies of 90.1%. According to the Jeffries–Matusita (J–M) distances, the short wavelength infrared band provides the greater contribution (the highest value was observed for band 6 in OLI data).     

Disaggregating and mapping crop statistics using hypertemporal remote sensing

Governments compile their agricultural statistics in tabular form by administrative area, which gives no clue to the exact locations where specific crops are actually grown. Such data are poorly suited for early warning and assessment of crop production. 10-Daily satellite image time series of Andalucia, Spain, acquired since 1998 by the SPOT Vegetation Instrument in combination with reported crop area statistics were used to produce the required crop maps. Firstly, the 10-daily (1998–2006) 1-km resolution SPOT-Vegetation NDVI-images were used to stratify the study area in 45 map units through an iterative unsupervised classification process. Each unit represents an NDVI-profile showing changes in vegetation greenness over time which is assumed to relate to the types of land cover and land use present. Secondly, the areas of NDVI-units and the reported cropped areas by municipality were used to disaggregate the crop statistics. Adjusted R-squares were 98.8% for rainfed wheat, 97.5% for rainfed sunflower, and 76.5% for barley. Relating statistical data on areas cropped by municipality with the NDVI-based unit map showed that the selected crops were significantly related to specific NDVI-based map units. Other NDVI-profiles did not relate to the studied crops and represented other types of land use or land cover. The results were validated by using primary field data. These data were collected by the Spanish government from 2001 to 2005 through grid sampling within agricultural areas; each grid (block) contains three 700 m × 700 m segments. The validation showed 68%, 31% and 23% variability explained (adjusted R-squares) between the three produced maps and the thousands of segment data. Mainly variability within the delineated NDVI-units caused relatively low values; the units are internally heterogeneous. Variability between units is properly captured. The maps must accordingly be considered “small scale maps”. These maps can be used to monitor crop performance of specific cropped areas because of using hypertemporal images. Early warning thus becomes more location and crop specific because of using hypertemporal remote sensing.     

Phenological metrics-based crop classification using HJ-1 CCD images and Landsat 8 imagery

Crop type data are an important piece of information for many applications in agriculture. Extracting crop type using remote sensing is not easy because multiple crops are usually planted into small parcels with limited availability of satellite images due to weather conditions. In this research, we aim at producing crop maps for areas with abundant rainfall and small-sized parcels by making full use of Landsat 8 and HJ-1 charge-coupled device (CCD) data. We masked out non-vegetation areas by using Landsat 8 images and then extracted a crop map from a long-term time-series of HJ-1 CCD satellite images acquired at 30-m spatial resolution and two-day temporal resolution. To increase accuracy, four key phenological metrics of crops were extracted from time-series Normalized Difference Vegetation Index curves plotted from the HJ-1 CCD images. These phenological metrics were used to further identify each of the crop types with less, but easier to access, ancillary field survey data. We used crop area data from the Jingzhou statistical yearbook and 5.8-m spatial resolution ZY-3 satellite images to perform an accuracy assessment. The results show that our classification accuracy was 92% when compared with the highly accurate but limited ZY-3 images and matched up to 80% to the statistical crop areas.     

Using high resolution QuickBird imagery for crop identification and area estimation

Imagery from recently launched high spatial resolution satellite sensors offers new opportunities for crop assessment and monitoring. A 2.8-m multispectral QuickBird image covering an intensively cropped area in south Texas was evaluated for crop identification and area estimation. Three reduced-resolution images with pixel sizes of 11.2 m, 19.6 m, and 30.8 m were also generated from the original image to simulate coarser resolution imagery from other satellite systems. Supervised classification techniques were used to classify the original image and the three aggregated images into five crop classes (grain sorghum, cotton, citrus, sugarcane, and melons) and five non-crop cover types (mixed herbaceous species, mixed brush, water bodies, wet areas, and dry soil/roads). The five non-crop classes in the 10-category classification maps were then merged as one class. The classification maps were filtered to remove the small inclusions of other classes within the dominant class. For accuracy assessment of the classification maps, crop fields were ground verified and field boundaries were digitized from the original image to determine reference field areas for the five crops. Overall accuracy for the unfiltered 2.8-m, 11.2-m, 19.6-m, and 30.8-m classification maps were 71.4, 76.9, 77.1, and 78.0%, respectively, while overall accuracy for the respective filtered classification maps were 83.6, 82.3, 79.8, and 78.5%. Although increase in pixel size improved overall accuracy for the unfiltered classification maps, the filtered 2.8-m classification map provided the best overall accuracy. Percentage area estimates based on the filtered 2.8-m classification map (34.3, 16.4, 2.3, 2.2, 8.0, and 36.8% for grain sorghum, cotton, citrus, sugarcane, melons, and non-crop, respectively) agreed well with estimates from the digitized polygon map (35.0, 17.9, 2.4, 2.1, 8.0, and 34.6% for the respective categories). These results indicate that QuickBird imagery can be a useful data source for identifying crop types and estimating crop areas.     

Textural Classification of Single Band SIR‐B Data over a Part of Brahmaputra Basin,India

Abstract

Texture is an important spatial feature, useful for identfying objects of regions of interest in an image. There are a number of methods for identification of textural parameters e.g. edgeness, frequency domain analysis, gray tone cooccurrence approach etc. Geologic information in radar images of heavily vegetated areas is contained mostly in the depiction of topography in image texture. Single band SIR/ERS‐1 SAR data posses a problem to the analyst for classification of the various land use/geological classes and generally multidate SAR data are used due to paucity of more number of bands. However, the multidate SAR data classification is not an ideal technique. A new technique, namely, spatial frequency band pass classification technique which generates two or more bands in the Fourier domain using the single band SAR data and then classifies various features of interest using their textural properties has been described here. Result shows higher percentage of classification using Maximum‐Likelihood Classifier (MXL) with two split‐band data as compared to the unsupervised classification of all the bands.     

Classification of Satellite Images in Assessing Urban Land Use Change Using Scale Optimization in Object-Oriented Processes (A Case Study: Ardabil City,Iran)

By using satellite imagery, the recognition and evaluation of various phenomena and extraction of information necessary for the planning of land resources or other purposes are easily accomplished. The purpose of this study is to compare the efficiency of seven commonly used methods of monitored classification of satellite data to evaluate land use changes using TM and OLI Landsat, IRS, Spot5 and Quick Bird bands as well as different color combinations of these images to detect agricultural land, residential areas and aquatic areas using object-oriented processing. Digital processing of satellite images was carried out in 1998 and 2016 using advanced methods. Training samples were extracted in five user classes by eCognition software using segmentation scale optimization, different color combinations and coefficients of shape and compression. The appropriate segmentation scale for arable land, human complications and the blue areas were, respectively, 50, 8 and 10. Then each image was classified separately using seven methods and extracted samples, and efficiency of each classification method was obtained by calculating two general health and Kappa coefficients. The results show that the accuracy of each classification method and the neural network with a total accuracy of 94.475 and Kappa coefficient of 92.095 were selected as the most accurate classification method. These results show that the sampling of educational samples with proper precision of the classes in the images and dependency probability of each satellite images pixel can be useful in classifying group available in helpful area.     

利用Landsat TM和IRS P6 LISS III卫星影像及GIS对耕地和耕种模式的时空动态分析(英文)

The study of the spatial patterns and temporal changes of cropland is important to understand the underlying factors and the functional effects of the agricultural landscape. On the other hand, crop dynamics mapping is essential to know the overall agro-spatial diversity of the area. Therefore, this paper addressed a spatio-temporal analysis of cropland and cropping pattern change in the Bogra district of Bangladesh over the last 16 years (between 1988/89 and 2004/05). In this paper, crop mapping from multi-temporal and multi-sensor satellite images was described. Landsat TM and IRS P6 LISS III satellite images were used with GIS for spatial dynamics of cropland and cropping pattern change analysis. First, seasonal cropland maps were derived from object-based classification of satellite images, then two-date classified image differencing with GIS overlay technique and decision rules were applied. Cropping pattern change was analyzed in a spatial and quantitative way for the 16 years and for this, Integrated Land and Water Information System (ILWIS) and Land Change Modular (LCM) of IDRISI Andes were used. The results showed that in the area, mono crop cultivation was found in summer, but in winter, areas under different crop cultivation had changed dramatically. Change analysis showed that the changes mainly occurred in the north northwest and southwest of the areas, and during the time the highest change area was found under the rice-potato pattern.      

Interpretation of crop growth patterns by means of NDVI‐time series in Zambia

A method to correlate crop production in Zambia to the yearly evolution of the Normalized Difference Vegetation Index (NDVI) is proposed. The method consists of the analysis of remote sensing data together with meteorological data and simulated crop production to obtain indicators of crop production. The accuracy of these indicators is assessed with statistical data.

The main objective was to assess whether the NDVI‐time series extracted from NOAA‐AVHRR‐images , having a pixel resolution of 73 km may give reliable information on crop production in Zambia where agricultural areas cover just 1% of the land area.

The mean NDVI‐value of several pixels, e.g. for one province or other administrative units, relates to the dominant type of vegetation in the area under consideration.

It is shown that the 7.3 km NDVI‐data give reliable indications on crop production in Zambia, when small areas (200–450 km² large ) are considered where agricultural land use is intensive. This implies that preliminary analysis is required to localize the agricultural areas. This has been done by means of high resolution satellite images i.e. LANDSAT‐MultiSpectral Scanner.

Consequently, the NDVI‐time series of the ‘agricultural ‘ pixels are used to calculate crop growth indicators which can be applied to assess the crop production.     

Mapping trees in high resolution imagery across large areas using locally variable thresholds guided by medium resolution tree maps

Large area tree maps, important for environmental monitoring and natural resource management, are often based on medium resolution satellite imagery. These data have difficulty in detecting trees in fragmented woodlands, and have significant omission errors in modified agricultural areas. High resolution imagery can better detect these trees, however, as most high resolution imagery is not normalised it is difficult to automate a tree classification method over large areas. The method developed here used an existing medium resolution map derived from either Landsat or SPOT5 satellite imagery to guide the classification of the high resolution imagery. It selected a spatially-variable threshold on the green band, calculated based on the spatially-variable percentage of trees in the existing map of tree cover. The green band proved more consistent at classifying trees across different images than several common band combinations. The method was tested on 0.5 m resolution imagery from airborne digital sensor (ADS) imagery across New South Wales (NSW), Australia using both Landsat and SPOT5 derived tree maps to guide the threshold selection. Accuracy was assessed across 6 large image mosaics revealing a more accurate result when the more accurate tree map from SPOT5 imagery was used. The resulting maps achieved an overall accuracy with 95% confidence intervals of 93% (90–95%), while the overall accuracy of the previous SPOT5 tree map was 87% (86–89%). The method reduced omission errors by mapping more scattered trees, although it did increase commission errors caused by dark pixels from water, building shadows, topographic shadows, and some soils and crops. The method allows trees to be automatically mapped at 5 m resolution from high resolution imagery, provided a medium resolution tree map already exists.     

Use of Principal Component Analysis (PCA) of Remote Sensing Images in Wetland Change Detection on the Kafue Flats,Zambia

Abstract

The paper describes the use of Principal Component Analysis (PCA) of remote sensing images as a method of change detection for the Kafue Flats, an inland wetland system in southern Zambia. The wetland is under human and natural pressures but is also an important wildlife habitat. A combination of Landsat MSS and TM images were used. The images used were from 24 September 1984 (MSS), 3 September 1988 (MSS), 12 September 1991 (TM) and 20 September 1994 (TM). They were geometrically co‐registered and, in the process, the 80m resolution MSS images were resampled to 30m using nearest neighbour resampling. Preliminary PCA revealed that for the MSS images most of the data variance was in near infrared reflectance while for the TM images it was in mid and thermal infrared bands. Holding sensor type constant, separate inter‐band correlation analysis for each image could indicate whether the wetland was drier or wetter on one date versus another. The 1994 image was made the reference image and equivalent green, red and near infrared bands from the other images were radiometrically normalised with those on the reference image. All the bands, three from each date, were then merged into a twelve‐band image on which PCA for change detection was undertaken. A colour composite of eigen images from the resulting principal components was used in change detection. Hydrological data, indicating long‐term reduced inflow of water into the wetland due to human regulation, help explain some of the wetland change detected. Compared to a classification comparison approach to change detection for this area, PCA was found to be very useful in indicating where change had occurred, though interpretation of the changes was difficult without reference to the input images. The methodology appears to have potential use in habitat monitoring for this wetland area.     

TM图像的光谱信息特征与最佳波段组合

本文分析了北自黑龙江省寒温带缓岗平原、南至广东省南亚热带丘陵等9个不同景观类型样区的TM图像数据,查明TM图像的光谱信息具3—4维结构,其物理含义相当于“亮度”、“绿度”和“热度”、“湿度”。在TM7个光谱图像中,一般以第5波段包含的地物信息最丰富。3个可见光波段(即第1,2,3波段)之间,两个中红外波段(即第5,7波段)之间相关性很高,表明这些波段的信息中有相当大的“重复性”或“冗余性”。第4,6波段则颇特殊,尤其是第4波段与各波段的相关性都很低,表明这个波段的信息有很大的独立性。计算20种组合的熵值的结果表明,由一个可见光波段、一个中红外波段及第4波段组合而成的彩色合成图像,一般具有最丰富的地物信息,其中又常以4,5,3或4,5,1波段的组合为最佳。     

Comparative evaluation of Indian remote sensing satellite and landsat thematic mapper data for geological and geomorphological applications

In order to evaluate the potentials of IRS‐1A Linear Imaging Self‐scanning Sensor (LISS‐I) data for geological and geomorphological applications and also to compare the IRS‐1A LISS‐I data with Landsat Thematic Mapper (TM) data, a study has been attempted for parts of Uttar Pradesh and Madhya Pradesh in Northern India. The first four spectral bands of Landsat TM sensor data which are similar and close to IRS‐1A LISS‐I senor have been utilised for the comparative evaluation. Various techniques employed for both the data set to derive the required geology and geomorphology related information include (i) band combination (ii) spectral response analysis (iii) principal component analysis (iv) supervised classification techniques and (v) visual observation of various outputs generated by the above methods. The Optimum Index Factor (OIF) method adopted for selecting suitable band combinations showed similar OIF rankings for IRS‐1A LISS‐I data and Landsat TM data. It has been visually observed that the band combination 1, 3 & 4 offers relatively better feature display. The spectral responses derived for various major geologic rock units such as Deccan Trap, Vindhyan Formation, Bundelkhand Granite and for a few landcovers such as surface water bodies and black soil show striking similarity in pattern for both LISS‐I and TM. The Principal Component (PC) analysis of both data sets suggested that the total scene brightness tends to dominate in the first PC. The percentage information contributed by PCs 1&2 as also by PCs 1,2 & 3 in both the LISS‐I and TM are comparable. It was observed from the classified image generated by performing supervised classification with a maximum likelihood algorithm that major geomorphic landforms were clearly distinguishable. Thus the qualitative and quantitative evaluation of both IRS‐1A LISS‐I and Landsat TM data showed that significant similarities exist between them. The study also revealed that IRS‐1A LISS‐I data can be effectively used for deriving geology and geomorphology related details.     

Discrimination of low-grade magnetite ores using remote sensing techniques

The remote sensing community in geology is widely using the Multispectral Landsat Thematic Mapper (TM) data which has a wider choice of spectral bands (six between 0.45 and 2.35 μm, plus a thermal infrared channel 10.4-12.5 urn). These were evaluated for low-grade magnetite ores mapping over the high-grade granulite region of Kanjamalai area of Tamil Nadu state, India. The Fourier Transform Infrared (FTIR) spectroscopy data (0.4-4.0 μm) for powders of the magnetite ores exposed with granulite rock and published spectral reflectance data were used as guides in selecting TM band reflectance ratios, which maximize discrimination of magnetite ores on the basis of their respective mineralogies. The study shows that the weathering mineralogy of magnetite ores causes absorption features in their reflectance spectra which are particularly characteristic of the near infrared. Comparison of TM data with field and petrographic observations shows the presence of magnetite and aluminosilicate minerals & show strong absorption at 0.7-1 μ.m wavelength spectral region & increase in the product of two TM band ratios: band 5 (1.55-1.75 μm) to band 4 (0.76-0.9 μm) and band 3 (0.63-0.69 μm) to band 4 (0.76-0.9 μm). Various computer image enhancement and data extraction techniques such as interactive digital image classification techniques using color compositing stretched ratio, maximum likelihood and thresholding statistical approaches using Landsat TM data are used to map the low-grade magnetite ores of the granulite region. The field traverses and local verification enhanced to map the other rock types namely granulites and gneisses of the study area.     

Delineation of surface waterlogged areas in parts of bihar using IRS-1C LISS-III data

Pre-monsoon and post-monsoon surface waterlogged areas were delineated using satellite remote sensing data for Muzaffarpur, Vaishali and Saran districts of North Bihar. Digital data of IRS-1C LISS-III sensor acquired on December 7, 1998 and April 6, 1999 were analyzed using digital image processing software-ERDAS Imagine 8.3.1. The surface waterlogged areas were delineated using modeling technique which is the most advanced and accurate method. Using the modeling technique, a pixel is classified as water if the digital number (DN) value of its Near Infra Red (NIR) band is less than the DN value of the Red band and the Green band, and the Normalized Difference Water Index (NDWI) is greater than or equal to 0.32. The pre-monsoon surface waterlogged areas are found to be 14.02, 23.61 and 9.61 km² while the post-monsoon surface waterlogged areas are found to be 231.83, 118.19 and 176.06 km² for Muzaffarpur, Vaishali and Saran districts, respectively. Also, land use/land cover maps were prepared.     

Study on the utility of IRS-P6 AWiFS SWIR band for crop discrimination and classification

This present study was conducted to find out the usefulness of SWIR (Short Wave Infra Red) band data in AWiFS (Advanced Wide Field Sensor) sensor of Resourcesat 1, for the discrimination of different Rabi season crops (rabi rice, groundnut and vegetables) and other vegetations of the undivided Cuttack district of Orissa state. Four dates multi-spectral AWiFS data during the period from 10 December 2003 to 2 May 2004 were used. The analysis was carried out using various multivariate statistics and classification approaches. Principal Component Analysis (PCA) and separability measures were used for selection of best bands for crop discrimination. The analysis showed that, for discrimination of the crops in the study area, NIR was found to be the best band, followed by SWIR and Red. The results of the supervised MXL classification showed that inclusion of SWIR band increased the overall accuracy and kappa coefficient. The ‘Three Band Ratio’ index, which incorporated Red, NIR and SWIR bands, showed improved discrimination in the multi-date dataset classification, compared to other SWIR based indices.     

AN EXPERIMENT IN AUTOMATED INTERPRETATION OF REMOTE SENSING IMAGERY

Algorithms, designed for digital image processing in standard mainframe computers and representing sequential stages in a land-use classification procedure, are used to produce maps of agricultural crop types from multispectral satellite imagery. Pixel reflectance values are first grouped according to an unsupervised “rapid classification algorithm,” or data compression procedure. Mean reflectance values of the resulting classes then go into a supervised “sequential clustering algorithm” where classes are refined according to training value and other parameter inputs. The objective is to increase the accessibility of automated image interpretation while balancing classification accuracy and processing time. Translated from: Vestnik Moskovskogo Universiteta, geografiya, 1984, No. 4, pp. 63-69.