This research develops a low cost remote sensing system for use in agricultural applications. The important features of the system are that it monitors the near infrared and it incorporates position and attitude measuring equipment allowing for geo-rectified images to be produced without the use of ground control points. The equipment is designed to be hand held and hence requires no structural modification to the aircraft. The portable remote sensing system consists of an inertia measurement unit (IMU), which is accelerometer based, a low-cost GPS device and a small format false colour composite digital camera. The total cost of producing such a system is below GBP 3000, which is far cheaper than equivalent existing systems. The design of the portable remote sensing device has eliminated bore sight misalignment errors from the direct geo-referencing process. A new processing technique has been introduced for the data obtained from these low-cost devices, and it is found that using this technique the image can be matched (overlaid) onto Ordnance Survey Master Maps at an accuracy compatible with precision agriculture requirements. The direct geo-referencing has also been improved by introducing an algorithm capable of correcting oblique images directly. This algorithm alters the pixels value, hence it is advised that image analysis is performed before image georectification. The drawback of this research is that the low-cost GPS device experienced bad checksum errors, which resulted in missing data. The Wide Area Augmented System (WAAS) correction could not be employed because the satellites could not be locked onto whilst flying. The best GPS data were obtained from the Garmin eTrex (15 m kinematic and 2 m static) instruments which have a highsensitivity receiver with good lock on capability. The limitation of this GPS device is the inability to effectively receive the P-Code wavelength, which is needed to gain the best accuracy when undertaking differential GPS processing. Pairing the carrier phase L1 with the pseudorange C/A-Code received, in order to determine the image coordinates by the differential technique, is still under investigation. To improve the position accuracy, it is recommended that a GPS base station should be established near the survey area, instead of using a permanent GPS base station established by the Ordnance Survey.

Since the early use of ground radar for precipitation detection in post-world war II, the radar has evolved on its own in precipitation remote sensing research and applications. The recent advances in radar remote sensing is, the development of polarimetric radar, also known as dual polarization radar, which has the capability of transmitting electromagnetic spectrums in both horizontal (H) and vertical (V) polarization states, thus providing additional information of the target precipitation particles by measuring polarimetric signatures, the reflectivity factor at H polarization (ZH) , differential reflectivity (ZDR) , differential propagation phase (ϕDP) , specific differential phase (KDP) , cross-correlation coefficient (PHV) and linear depolarization ratio (LDR). In commensurate with new era in precipitation remote sensing, this thesis explores the potential of polarimetric radar on the improvements in precipitation remote sensing in the UK context. All major area of the improvements aided by the polarimetry and polarimetric signatures are addressed. These include the clutter and anomalous propagation identification, attenuation signal correction, polarimetric rainfall estimators, drop size distribution retrievals, bright band/melting layer recognition and hydrometeor classification. Several novel approaches and investigations dealing with the polarimetric improvements are scrutinized and proposed in terms of numerical analysis, while some of them employ artificial intelligence (AI) techniques. Key original contributions in synergy with polarimetric radar signatures on precipitation remote sensing are: 1) long-term disdrometer DSD analysis to support the development of polarimetry based algorithms and models, 2) the use of several AI techniques such as support vector machine, artificial neural network, decision tree, and nearest neighbour system for clutter identification, 3) the sensitivity a:ssociated with total differential propagation phase constraint (ΔϕDP) on ZH correction for attenuation, 4) the exploration of polarimetric rainfall estimators [R(ZH, ZDR, Knp)] for rainfall estimation, 5) a genetic programming approach for drop size distribution retrievals [Do(ZH, ZDR) , Nw(ZH, ZDR, Do), μ(ZH, ZDR, Do)], and its use for convective/stratiform rain indexing, and 6) a fuzzy logic based system for automatic melting layer/bright band recognition and hydrometeor classification as well as appraisal with a numerical weather prediction (NWP) model and radio soundings observations. In fact, the radar polarimetry has been proven not only to improve data quality and precipitation estimation, but also characterizing the precipitation particles, thus has a great potential on fostering the precipitation remote sensing research and applications. Keywords: polarimetric radar; dual polarization radar; microphysics of precipitation; drop size distribution (DSD); clutter and anomalous propagation identification; attenuation correction; rainfall estimators; microphysical DSD retrievals; melting layer and bright band detection; hydrometeor classification.

Remote sensing is a quick and inexpensive way of gathering information about the Earth. It enables one to constantly get updated information from satellite images for real-time local and global mapping of environmental changes. Current classification methods used for extracting relevant knowledge from this huge information pool are not very efficient because of the limited training samples and high dimensionality of the images. Information fusion is often used in order to improve the classification accuracy prior or after performing classification. However, these techniques cannot always successfully overcome the aforementioned issues. Therefore, in this thesis, new methods are introduced in order to increase the classification accuracy of remotely sensed data by means of information fusion techniques. This thesis is structured in three parts. In the first part, a novel pixel based image fusion technique is introduced to fuse optical and SAR image data in order to increase classification accuracy. Fused images obtained via conventional fusion methods may not contain enough information for subsequent processing such as classification or feature extraction. The proposed method aims to keep the maximum contextual and spatial information from the source data by exploiting the relationship between spatial domain cumulants and wavelet domain cumulants. The novelty of the method consists in integrating the relationship between spatial and wavelet domain cumulants of the source images into an image fusion process as well as in employing these wavelet cumulants for optimisation of weights in a Cauchy convolution based image fusion scheme. In the second part, a novel feature based image fusion method is proposed in order to increase the classification accuracy of hyperspectral images. An application of Empirical Mode Decomposition (EMD) to wavelet based dimensionality reduction is presented with an aim to generate the smallest set I of features that leads to better classification accuracy compared to single tech! niques. Useful spectral information for hyperspectral image classi6cation can be oj:>tained by applying the Wavelet Transform (WT) to each hyperspectral signature. As EMD has the ability to describe short term spatial changes in frequencies, it helps to get a better understanding of the spatial information of the signal. In order to take advantage of both spectral and spatial information, a novel dimensionality reduction method is introduced, which relies on using the wavelet transform of EMD features. This leads to better class separability and hence to better classification.

There is a growing need for measurements of surface fluxes which are rep- resentative of large, heterogenous surfaces, for example urban areas. This requires an instrument which makes measurements that have a large source area, but the source area of measurements made by traditional, surface- based in situ instruments are generally small compared to the surface area required. A solution to this problem is to use a remote sensing instru- ment, such as a Doppler lidar, which is able to make measurements with a potential source area of the order of 10s of km2. We describe a novel technique, in which measurements of turbulence made by a Doppler lidar in the convectively unstable daytime boundary layer are used with an optimal inverse method to make estimates of surface sensi- ble heat fluxes with a large source area. The optimal inverse method uses forward models which are based upon functions of the vertical velocity vari- ance derived from mixed layer similarity scaling theory. The optimal inverse method also provides a probabilistic error of the estimated heat flux, which incorporates both instrument and sampling errors. Testing of the optimal inverse method upon a simulated convective bound- ary layer generated by an LES provides confidence in the method, and also allows for a critical examination of the similarity functions used in the for- ward models. The optimal inverse method is then applied to a set of lidar data from the Chilbolton observatory, and the estimated heat fluxes are compared to those measured by a sonic anemometer at the same site. The results are correlated, but show a bias in which the heat fluxes estimated by the optimal inverse method are consistently lower than those estimated by the sonic anemometer. Possible reasons for this bias are explored; in partic- ular, the bias is larger in the morning when the lidar measurements appear to show that there is shear turbulence present in the transitional convective layer, rendering the forward models inappropriate. The lidar also measures large spikes of turbulence in the middle of the boundary layer which appear during the afternoon. The reasons for these phenomena are unclear, and further work is required to investigate whether or not they are site specific.

Remote sensing satellites are often built with payloads that do not include line of sight steering mechanisms, such that pointing their payloads requires rotation of the whole satellite. In cases, when frequent line of sight retargeting is required, there is a need for efficient actuators and control schemes that would support rapid attitude maneuvering together with adequate pointing accuracy and stability between the maneuvers. These control schemes shall accommodate a variety of realistic conditions, such as general three dimensional maneuver direction, existence of initial and/or final angular rates, non zero net angular momentum and various actuators constraints. Within this frame, this research develops the Backwards Planning approach as one of the possible control methods for rapid maneuvering. The method is based on state feedback and combines time efficiency together with straight forward computation flow. Novel efficient methods to execute the Backwards Planning Control in the 3D attitude space are proposed here. The methods refer both for the first saturated control phase of the maneuver and for the last braking phase. The actuators used for the spacecraft control in this research are either Reaction Wheels (RWs) or Single Gimbal Control Moment Gyros (SGCMGs) or both of them together. The advantage of the SGCMG is in rapid rotational maneuvering, but their application for high quality pointing requires very accurate gimbal mechanisms. On the other hand, RWs are usually more suitable for accurate pointing, but their torque to power performance is inferior in maneuvering. It is shown that the coordination of SGCMGs and RWs together enables to draw more performance from the SGCMGs in terms of agility and meet the pointing requirements between maneuvers where only the RWs are used. Novel SGCMG steering laws are suggested as well. While the steering laws determine the required angular rate for each gimbal, most steering laws are defined in the angular momentum domain and output the gimbals angular rates to produce a given required torque or angular momentum increment. This research however, practices a novel steering law in the gimbal angles domain. While both steering laws turn to be dynamically equivalent for small control signals, as in the steady state, it is shown that the steering in the gimbal angles domain is more effective in maneuvering with the Backwards Planning control logic.

The first part of this study evaluates whether video can be used to provide the information required for pipeline ROW monitoring. Comparisons of information content are made between the aerial photograph (as a representative of typical wide angel sensors) and video. The aerial photographic image had neither the ground pixel size nor inexpensive data within the narrow target due to its wide angle-of-view. The video realistically isolated the major communities of the narrow pipeline ROW by reliable spatial precision due to its narrow angle-of-view. These investigations led to the conclusions that video is the sensor, technically and economically, which can meet the information requirements for the proposed target. The second part of the study develops a digital mosaicking procedure for narrow strip video. Until now, due to the time consuming mosaic requirement, video imagery had tended to be neglected in a digital environment. A reliable mosaic creation method for strip video was developed by incorporating traditional analogue mosaicking and digital image processing. An interactive approach, registering video frames bi-directionally, produced an acceptable positional accuracy. The concept of bi-directional bridging is a major product of this project. Bi-directional bridging, requiring solely end lap, enables VSM to be a powerful remote sensing tool in terms of time and labour. The development of bi-directional bridging breaks down the typical concept of video as being used purely as a snapshot visual assessment tool. The final aspect of the project is related to change-detection of the pipeline ROW recovery. A definite requirement in post-construction management of the ROW site is to detect recovery status on a multi-temporal basis. However, such large-scale video systems are often discussed as being inadequate for a change-detection application due to geometric and radiometric calibration problems. In spite of such limitations, changes to several detailed land cover classes, particularly for the rapid recovered target of the ROW (usually 5-10 years), could be detected successfully by visual or quantitative methods and through further patch dynamics analysis in a GIS environment. The results of this study indicate that such calibration problems are generally not a major drawback in acquiring change-detection information in a practical operational application which requires mostly generalised thematic mapping for relatively simple classes.

This report quantifies the motion of wheat subject to wind and assesses the effect of this motion on the coherence obtained from Synthetic Aperture Radar (SAR) interferometry. Over vegetation, the loss of coherence due to the change in backscatter between two SAR images taken at a different time (temporal decorrelation) is related to the wind induced motion of vegetation elements. The research aims to provide simultaneous in situ measurements of crop motion and wind velocity at canopy height and to use these measurements in a coherence model to determine the quantitatively the parameters which infer temporal decorrelation. The potential of coherence for agricultural applications is assessed. The three-dimensional motion of wheat is measured by a photogrammetry method using two commercially available video cameras. Simultaneously, wind velocity at canopy height is measured by anemometers at a high sampling frequency. Wheat motion and wind velocity data were collected in a field local to Cranfield University in summer 2000. The CD attached to this report contains the wheat motion and wind velocity data. They show that the motion of wheat is correlated with the wind speed, and that wheat plants adjacent to each other move coherently. The coherence model is based on a statistical approach, which represents the total backscatter from vegetation as the phasor addition of a fixed component and one or more components which are weather dependent. The relative contributions of the total backscatter are estimated with the RT2 backscatter intensity model. The motion measurements are used to define the variability of the phase of the weather dependent components in the model. Outputs of the model show that a C-band SAR with an incidence angle of 23° (typical configuration of the ERS satellites) yields coherence values highly variable with the wind conditions at the time of the radar passes. The potential use of coherence for agricultural applications is limited by this variability, which infers the need for an accurate coherent backscatter model.

A low cost system for measuring insect radar cross section (RCS) as a function of polarisation orientation when viewed at constant vertical aspect is presented. A low power continuous wave system (frequency 9.4 GHz) was developed which illuminated a target vertically and measured the power reflected back as the target was rotated in the horizontal plane. Data was collected by a microcomputer and stored on floppy disk for later analysis. A standard target was measured before each insect target and software used this data to calibrate the system for each insect target measurement. A 5 parameter mathematical model, based on the scattering matrix, is described. Software was developed to calculate the parameters from collected data using a least squares procedure. Measurements from 54 specimens representing 18 species of commonly available insect are reported. Average RCS’s were similar to the RCS of water spheres of equivalent mass. For small insects, the maximum and minimum RCS occurred when the electric vector (E-vector) was parallel and perpendicular to the body axis respectively. As insect size increased, a subsidiary maximum developed when the E-vector was perpendicular to the body axis, becoming dominant for the largest insects measured. The shapes of the RCS pattern were approximately ‘mirror’ symmetric, except for the RCS curves from S. gregaria, the largest insects measured, which were asymmetrical and sensitive to the position of the large rear legs. It was found that the ratio of maximum RCS to minimum RCS bore no simple relationship to the ratio of body length to abdomen width of the insect, even when adjusted for mass and was not a useful measure of target shape. The implications of this data are discussed and suggestions for further work are presented.

In disaster monitoring, an on-board autonomous system, which can process the images to extract useful images is considered necessary to speed up the decision making process. However, the current degree of automation in image processing on-board remote sensing satellites is low. Most of the remote sensing satellites operate a 'store-and-forward' mechanism and do not have intelligent imaging capabilities. This is because of the satellites being equipped with low-performance computing capabilities to save power and due to reliability considerations. Also, the existing algorithms for disaster monitoring are not fully automated and require high-performance computing. The objective of this research project is to investigate the feasibility of implementing an intelligent system on board small satellites that can perform critical image processing tasks such as image registration and change detection. Investigation of current change detection and flood detection techniques is carried out in order to evaluate their performance against state-of-the-art on-board computing systems of small satellites. Other than the limited computing capabilities and possible errors that might occur due to the radiation in space; image registration, change detection and flood detection are very challenging tasks to be implemented on board small satellites, because these tasks need to process a pair of images (i.e. comparing them): sensed and reference images. This thesis proposes a robust change detection framework to reduce the problems of pixel-to-pixel comparison. Image tiling and fuzzy inference engine are introduced in the system as a method to overcome the problems caused by direct pixel-to-pixel comparison. The performance results show that the proposed method has a better performance than pixel-to-pixel methods. The proposed new on-board flood monitoring system is simulated to evaluate its performance and to verify if it is viable for the small satellite implementation. GPS reflectometry data is proposed to be used together with existing multispectral images to reduce the problem caused by cloud cover. Recommendations and requirements for a small satellite mission for monitoring flood are presented.

The primary aims and objectives of this thesis are to identify the sources and operation of the errors which are introduced during end-user post-processing of airborne laser scanning data. Previous research has concentrated on the errors incorporated during data capture and preliminary supplier processing. The errors which are introduced by the end-users have been largely neglected. As a result, data users cannot currently estimate the errors within, and therefore the quality of, the models they produce. Laser scanning is a remote sensing technique for the capture of height data of the surface of the Earth. It offers competitive capture costs, high accuracy, and is particularly suited to capturing information in complex urban areas. As a result the commercial value of laser scanning data is high. However, in order to realise the potential of this technique, the quality of the datasets derived from the data must be assessed and the errors introduced during modelling understood. For users to make informed decisions regarding the design of their post-processing workflow it is fundamental that they know how and where errors may be introduced. The characteristics of these errors are investigated in this thesis using a range of approaches. End-user post-processing is divided into three techniques in the thesis: data structuring, filtering and segmentation. Each process is investigated hi terms of accuracy and sensitivity, through the comparison of several methods with reference models. New algorithms for filtering and segmenting laser data are presented. The errors created by each process are identified and analysed. The location of errors across the elevation surface are also investigated. It is shown how this information could be used to aid end-users design their post-processing methodology. The methodology for analyzing the errors is presented as a framework which could be used as a standard for ALS models. This thesis shows that the choice of post-processing methodology can significantly alter both the magnitude and spatial pattern of errors with a model derived from airborne laser scanning data. The differences between modeling strategies, and the importance of these differences, is shown with reference to a flood modeling application. Finally, strategies for minimizing error for post-processing are proposed.