A Systematic Literature Review of Machine Learning and Deep Learning Approaches for Spectral Image Classification in Agricultural Applications Using Aerial Photography

Recently,there has been a notable surge of interest in scientific research regarding spectral images.The potential of these images to revolutionize the digital photography industry,like aerial photography through Unmanned Aerial Vehicles(UAVs),has captured considerable attention.One encouraging aspect is their combination with machine learning and deep learning algorithms,which have demonstrated remarkable outcomes in image classification.As a result of this powerful amalgamation,the adoption of spectral images has experienced exponential growth across various domains,with agriculture being one of the prominent beneficiaries.This paper presents an extensive survey encompassing multispectral and hyperspectral images,focusing on their applications for classification challenges in diverse agricultural areas,including plants,grains,fruits,and vegetables.By meticulously examining primary studies,we delve into the specific agricultural domains where multispectral and hyperspectral images have found practical use.Additionally,our attention is directed towards utilizing machine learning techniques for effectively classifying hyperspectral images within the agricultural context.The findings of our investigation reveal that deep learning and support vector machines have emerged as widely employed methods for hyperspectral image classification in agriculture.Nevertheless,we also shed light on the various issues and limitations of working with spectral images.This comprehensive analysis aims to provide valuable insights into the current state of spectral imaging in agriculture and its potential for future advancements.

Robust key point descriptor for multi-spectral image matching

Histogram of collinear gradient-enhanced coding （HCGEC）, a robust key point descriptor for multi-spectral image matching, is proposed. The HCGEC mainly encodes rough structures within an image and suppresses detailed textural information, which is desirable in multi-spectral image matching. Experiments on two multi-spectral data sets demonstrate that the proposed descriptor can yield significantly better results than some state-of- the-art descriptors.

Evaluation of Structural Patterns and Related Alteration and Mineralization Zones by Using ASAR-ASTER Imagery in Siyahrood Area (East Azarbaijan—NW Iran)

The NW part of Iran belongs to the Iranian plateau that is a tectonically active region within the Alpine-Himalayan orogenic belt. The intrusion of Oligocene parts in various faces caused the alteration and mineralization such as copper, molybdenum, gold and iron in the Siyahrood area. Granitoidic rocks with component of Granodiorite to alkali have been influenced by hydrothermal fluids. Alteration zones are important features for the exploration of deposits and the ASTER sensor is able to identify the type of alteration and its alteration zoning. This method can be a useful tool for detecting potential mineralization area in East Azarbaijan—Northwest of Iran. The purpose of this study is to evaluate ASTER data for mapping altered minerals in Siyahrood area in order to detect the potential mineralized areas. In this study, false color composite, and band ratio techniques were applied on ASTER data and argillic, phyllic, Iron oxide and propylitic alteration zones were separated. ASAR image processing has been used for lineaments and faults identified by the aid of directional filter. The structural study focused on fracture zones and their characteristics including strike, length, and relationship with alteration zones. The results of this study demonstrate the usefulness of remote sensing methods and ASTER multi-spectral data for alteration, and ASAR data are useful for lineament mapping.

Lineament Patterns and Mineralization Related to Alteration Zone by Using ASAR-ASTER Imagery in Hize Jan-Sharaf Abad Au-Ag Epithermal Mineralized Zone (East Azarbaijan—NW Iran)

East Azarbaijan belongs to the Iranian plateau and is part of lesser Caucasus province. Studied area is located in west-central Alborz. The intrusion of oligocene bodies in various units causes the alteration and mineralization in northwest of Iran. The Hizejan-Sharafabad is one of this named mineralized zone. Granitoidicrocks with component of Granodiorite to alkali have been influenced by hydrothermal fluids. Fractures and faults are as weak zone in earth surface and hydrothermal fluids rise to surface by these geological structures. These solutions cause to alteration in host rocks. Alteration zones are important features for the exploration of deposits. The altered rocks have specific absorption in some spectral portion and ASTER sensor is able to identify the type of alteration. Remote sensing method is useful tool for discovering altered area. The purpose of this study is to appraise ASTER data for surveying altered minerals in Hizejan-Sharafabad area in the event of detecting the potential mineralized areas. In this research, False Color Composite (FCC), Band ratio, and color composite ratio techniques are applied on ASTER data and Silica, Argilic, and Propylitic alteration zones are detected. These alteration types and mineralized area are related to Hizejan–Sharafabad fault which is absent in the fault maps. ASAR image processing has been used for lineaments and faults identified by the aid of Directional and Canny Algorithm filters. The structural study focuses on fracture zones and their characteristics including strike, length, and relationship with alteration zones.

Application of PCA Numalgorithm in Remote Sensing Image Processing

A numerical algorithm of principal component analysis (PCA) is proposed and its application in remote sensing image processing is introduced: (1) Multispectral image compression;(2) Multi-spectral image noise cancellation;(3) Information fusion of multi-spectral images and spot panchromatic images. The software experiments verify and evaluate the effectiveness and accuracy of the proposed algorithm.

Application of the Dempster-Shafer Theory to the Classification of Pixels from Aster Satellite Images and Spectral Indices

In this paper, it is proposed to apply the Dempster-Shafer Theory (DST) or the theory of evidence to map vegetation, aquatic and mineral surfaces with a view to detecting potential areas of observation of outcrops of geological formations (rocks, breastplates, regolith, etc.). The proposed approach consists in aggregating information by using the DST. From pretreated Aster satellite images (geo-referencing, geometric correction and resampling at 15 m), new channels were produced by determining the spectral indices NDVI, MNDWI and NDBaI. Then, the DST formalism was modeled and generated under the MATLAB software, an image segmented into six classes including three absolute classes (E,V,M) and three classes of confusion ({E,V}, {M,V}, {E,M}). The control on the land, based on geographic coordinates of pixels of different classes on said image, has made it possible to make a concordant interpretation thereof. Our contribution lies in taking into account imperfections (inaccuracies and uncertainties) related to source information by using mass functions based on a simple support model (two focal elements: the discernment framework and the potential set of belonging of the pixel to be classified) with a normal law for the good management of these.

Multi-Spectral and Fluorescence Imaging in Prevention of Overdose of Herbicides: The Case of Maize

Evaluation of the impact of herbicides on maize was done through multi- spectral and multi-modal imaging and multi-spectral fluorescence imaging combined with statistical methods. Spectra containing 13 wavelengths ranging from 375 nm to 940 nm were derived from multi-spectral images in transmission, reflection and scattering mode and fluorescence images obtained using high-pass filters (F450 nm, F500 nm, F550 nm, F600 nm, F650 nm) on control maize samples and maize samples treated with Herbextra herbicide were used. The appearance of the spectra allowed us to characterize the effect of the herbicide on the maize pigment concentration. The fluorescence images allowed us to track the fate of absorbed energy and through PLS-DA and SVM-DA to discriminate the two leaf categories with very low error rates for the test, i.e. 4.9% and 2% respectively. The results of this technique can be used in the context of precision agriculture.

Identifying ephemeral gullies from high-resolution images and DEMs using flow-directional detection

Ephemeral gullies,which are widely developed worldwide and threaten farmlands,have aroused a growing concern.Identifying and mapping gullies are generally considered prerequisites of gully erosion assessment.However,ephemeral gully mapping remains a challenge.In this study,we proposed a flow-directional detection for identifying ephemeral gullies from high-resolution images and digital elevation models(DEMs).Ephemeral gullies exhibit clear linear features in high-resolution images.An edge detection operator was initially used to identify linear features from high-resolution images.Then,according to gully erosion mechanism,the flow-directional detection was designed.Edge images obtained from edge detection and flow directions obtained from DEMs were used to implement the flow-directional detection that detects ephemeral gullies along the flow direction.Results from ten study areas in the Loess Plateau of China showed that ranges of precision,recall,and Fmeasure are 6 o.66%-90.47%,65.74%-94.98%,and63.10%-91.93%,respectively.The proposed method is flexible and can be used with various images and DEMs.However,analysis of the effect of DEM resolution and accuracy showed that DEM resolution only demonstrates a minor effect on the detection results.Conversely,DEM accuracy influences the detection result and is more important than the DEM resolution.The worse the vertical accuracy of DEM,the lower the performance of the flow-directional detection will be.This work is beneficial to research related to monitoring gully erosion and assessing soil loss.

USING COVARIANCE INTERSECTION FOR CHANGE DETECTION IN REMOTE SENSING IMAGES

In this paper,an unsupervised change detection technique for remote sensing images acquired on the same geographical area but at different time instances is proposed by conducting Covariance Intersection(CI) to perform unsupervised fusion of the final fuzzy partition matrices from the Fuzzy C-Means(FCM) clustering for the feature space by applying compressed sampling to the given remote sensing images.The proposed approach exploits a CI-based data fusion of the membership function matrices,which are obtained by taking the Fuzzy C-Means(FCM) clustering of the frequency-domain feature vectors and spatial-domain feature vectors,aimed at enhancing the unsupervised change detection performance.Compressed sampling is performed to realize the image local feature sampling,which is a signal acquisition framework based on the revelation that a small collection of linear projections of a sparse signal contains enough information for stable recovery.The experimental results demonstrate that the proposed algorithm has a good change detection results and also performs quite well on denoising purpose.

Modeling and Characterization of Vegetation, Aquatic and Mineral Surfaces Using the Theory of Plausible and Paradoxical Reasoning from Satellite Images: Case of the Toumodi-Yamoussoukro-Tiébissou Zone in V Baoulé(Côte d’Ivoire)

In this paper, the theory of plausible and paradoxical reasoning of Dezert- Smarandache (DSmT) is used to take into account the paradoxical charac-ter through the intersections of vegetation, aquatic and mineral surfaces. In order to do this, we developed a classification model of pixels by aggregating information using the DSmT theory based on the PCR5 rule using the ∩NDVI, ∩MNDWI and ∩NDBaI spectral indices obtained from the ASTER satellite images. On the qualitative level, the model produced three simple classes for certain knowledge (E, V, M) and eight composite classes including two union classes characterizing partial ignorance ({E,V}, {M,V}) and six classes of intersection of which three classes of simple intersection (E∩V, M∩V, E∩M) and three classes of composite intersection (E∩{M,V}, M∩{E,V}, V∩{E,M}), which represent paradoxes. This model was validated with an average rate of 93.34% for the well-classified pixels and a compliance rate of the entities in the field of 96.37%. Thus, the model 1 retained provides 84.98% for the simple classes against 15.02% for the composite classes.

ASTER卫星遥感影像在冬小麦品质监测方面的初步应用

该文利用A STER遥感数据在冬小麦品质监测方面进行了初步的应用。根据2004年搭载A STER传感器的T erra卫星的过境周期,对北京郊区21个冬小麦试验点进行田间取样试验,通过分析试验点冬小麦叶片生化组分(叶绿素a)与氮素间的关系、冬小麦叶片氮素含量与籽粒品质(蛋白质含量)间的关系、以及冬小麦A STER影像冠层光谱信息与叶片生化组分和籽粒品质之间的关系,发现:冬小麦灌浆期小麦叶片叶绿素含量与同期A STER影像的光谱信息-NDV I之间有良好的相关性,而冬小麦籽粒蛋白质含量也与A STER影像波段2之间有很好的相关性;利用A STER遥感影像对冬小麦灌浆期叶绿素以及籽粒蛋白质含量进行建模并反演,能够在较大的范围里了解冬小麦的品质信息,并能够在一定程度上对作物的籽粒蛋白质含量等品质进行预测;研究还表明,利用遥感技术来进行小麦籽粒蛋白质等主要品质指标的预测是可行的。研究成果为利用遥感技术监测冬小麦长势和籽粒品质提供了理论依据,开辟了遥感应用的新途径。

利用ASTER遥感资料提取南京城郊土地利用信息的研究

土地资源是社会经济发展最基本的物质基础,利用遥感技术可对土地资源进行分类调查。在对地观测方面,高光谱遥感与一般遥感影像相比,能提供更高质量的信息源。先进的空间热辐射反射计(ASTER)覆盖可见光、近红外、短波红外和热红外的14个波段,分辨率从15m到90m,对长江流域环境监测、地表覆盖变化、短期气候预测、自然灾害监测等,对于了解该区资源、生态与环境具有重要的意义。该文利用2001年ASTER数据对南京城郊地区进行了土地利用分类,首先采用多元统计方法提取分类特征波段。然后以非监督分类获得初始训练样本,对训练区进行多次像元提纯后,运用监督分类方法对南京城郊土地利用信息进行分类。研究表明,多种分类方法相结合可从ASTER影像更好地提取土地利用信息,达到较好得分类精度。

ASTER影像在喀斯特石漠化遥感解译中的应用——以贵州省瓮安县为例

以2004年ASTER影像数据为信息源,在GPS野外调研和G IS软件支持下,对贵州省瓮安县喀斯特石漠化进行遥感解译,结果表明ASTER高光谱遥感与一般遥感影像相比能为对地观测提供更高质量的信息源,能较好地满足1∶5万石漠化成图的需要,为制定喀斯特石漠化综合防治规划提供依据。

ASTER影像提取植被信息的NDVI与SAVI法比较——以广州花都区为例

基于ASTER遥感影像,使用IDL语言编写归一化差异植被指数(NDVI)和土壤调整植被指数(SAVI)的计算公式,对遥感影像进行处理,分别对两种方法处理后的遥感影像采用K-Means分类,经过分类后处理,提取植被信息。NDVI整体上较好地反映了不同土地覆被信息;而SAVI对于各种地类的值域较宽,反映绿色植被内部差异信息较明显,可为不同植被类型的信息提取提供方法参考。

多时相ASTER遥感影像相对辐射校正研究

遥感影像获取时受传感器本身、太阳高度角、大气、地形等因素对地面光谱反射信号的影响,导致多时相影像上同一目标的光谱特征差异很大.为了进一步的应用研究,必须要先进行辐射校正.以黑龙江省海伦市及其周边地区作为研究区,将太阳高度角校正、辐射校正、日地距离校正和相对辐射校正方法结合使用,将其它因素引起的光谱值差异减小到最小值,在一定程度上提高了校正精度,达到预期目的.

ASTER数据在北方铝土矿预普查中的应用——以豫西渑池地区为例

ASTER数据的光谱和空间分辨率均具有一定优势,适合于中等比例尺的矿产预普查。为了探讨ASTER数据应用于铝土矿预普查的效果,通过对ASTER图像彩色合成、地质解译及异常信息提取,对豫西渑池地区的铝土矿进行了预普查试验。结果表明,在北方基岩裸露地区,应用ASTER数据圈定"赋矿影像单元"和提取铝土矿遥感找矿信息能获得较好的效果。

基于ASTER影像数据源的龙川江LUCC景观格局特点分析

以高级星载热发射反照辐射计(ASTER)影像作为数据源,以龙川江流域作为研究对象,在遥感和GIS技术的支持下,采用GIS和景观生态学相结合的办法,对龙川江流域土地利用/覆被景观空间格局特点进行分析。结果表明,龙川江流域植被覆盖率较高,林地和灌草丛面积合计占研究区总面积的86.27%,但除灌草丛连通性较强外,各景观类型斑块小而分布零散,景观破碎化程度大,阻碍了整个生态系统的正常的物质和能量流动,生态环境需要改善。

基于TM和ASTER影像的福州市土地利用变化分析

20世纪80年代以来，福州市作为我国首批14个沿海开放城市之一，经历了快速发展时期，其土地利用发生了巨大变化．研究较长时期区域土地利用的时空变化，需要多源遥感数据结合应用．文章基于TM和ASTER遥感影像，提取福州市1988～2004年16年的土地利用变化信息，分析土地利用变化的数量、速率和空间特征．结果表明，福州市在这16年间耕地面积比基期减少了33％，建设用地比基期增加了88％，土地利用类型转换的空间特征反映地貌条件和城市扩展策略的共同制约．研究采用多源遥感数据结合应用的方法，有助于延长区域土地利用变化监测数据的时间序列，可供同类研究借鉴．

ETM＋与ASTER数据在老挝红土型铝土矿勘查中的应用

老挝南部波罗芬高原红土型铝土矿大面积直接裸露地表。在铝土矿勘查过程中,利用ETM+和ASTER多光谱遥感数据,分别采用"比值+主成分分析+监督分类"和"比值+主成分分析+光谱角填图"两种不同的方法提取了红土型铝土矿矿化蚀变信息,并利用所提取的矿化信息直接圈定铝土矿矿体,大大提高铝土矿勘查效率,省时省力,为同类型的铝土矿和红土型镍矿的勘查提供了一种快速有效经济的方法和途径。