This thesis addresses the issue of ground collision in poor weather conditions. As bad weather is an adverse factor when airplanes are taxiing, an obstacle detection system based on thermal vision is proposed to enhance the awareness of pilots during taxiing in poor weather conditions. Two infrared cameras are employed to detect the objects and estimate the distance of the obstacle. The distance is computed by stereo vision technology. A warning will be given if the distance is less than the safe distance predefined. To make the system independent, the proposed system is an on-board system which does not rely on airports or other airplanes. The type of obstacle is classified by the temperature of the object. Fuzzy logic is employed in the classification. Obstacles are classified into three main categories: aircraft, vehicle and people. Membership functions are built based on the temperature distribution of obstacles measured at the airport. In order to improve the accuracy of classification, a concept of using position information is proposed. Different types of obstacle are predefined according to different area at the airport. In the classification, obstacles are classified according to the types limited in that area. Due to the limitation of the thermal infrared camera borrowed, images were captured first and then processed offline. Experiments were carried out to evaluate the detecting distance error and the performance of system in poor weather conditions. The classification of obstacle is simulated with real thermal images and pseudo position information at the airport. The results suggest that the stereo vision system developed in this research was able to detect the obstacle and estimate the distance. The classification method classified the obstacles to a certain extent. Therefore, the proposed system can improve safety of aircraft and enhance situational awareness of pilots. The programming language of the system is Python 2.7. Computer graphic library OpenCV 2.3 is used in processing images. MATLAB is used in the simulation of obstacle classification.

This work deals with the measurement of temperature fields in the air using an infrared camera. The dissertation describes the opportunity of measuring the temperature field in the air and the characterization of the developed measuring method. In the next part there are introduced the beginning of thermography imaging and the field of usability of the new infrared measuring method. Further, the theoretical foundations of the thermography measuring method in the temperature fields in the air are described. In the theoretic background there are described the analysis of heat conduction in an auxiliary material, the determination of dynamic properties of the method and the analysis of radiation, which has an influence on infrared imagining. This method requires an insertion of the auxiliary material into the non-isothermal air flow, which can allow for the study of the temperature distribution in air. For effective visualization of temperature fields in the air using an infrared camera, the selection of the appropriate auxiliary material, on which the air temperature displays, is crucial. In the next part of the doctoral thesis, there is a description of static measuring properties of auxiliary materials. The usability range of the measuring method is determined from these properties. In the thesis there are presented the description of the device for the measurement of 2D temperature fields in the air and the description of the measuring device for 3D measurements of temperature fields in the air using an infrared camera, which can also be used for measurements of temperature fields in small enclosed spaces through a viewing window. For the practical use of the method, the detailed methodology of measuring temperature fields in the air by an infrared camera was developed and its applicability was demonstrated on practice examples. The developed measuring method can be used in many areas of research and in practice.

none

In this thesis work, temperature estimation algorithms based on physical and radiometric approaches are developed. Developed algorithms, firstly, tested on artificial images for different test cases. Following this, algorithms are tried out on real infrared images in order to verify that they are working properly. Finally, temperature estimations are done by including emissivity. Obtained results are compared to the temperature estimation results of a reference infrared camera. All the results and errors obtained during this study are presented and discussed.

This Master's theses deals with the contactless measurement of surface temperatures and incorporates it among other methods of temperature measurement. For practical experiments were used infrared camera ThermaCAM SC 2000 from producer FLIR. The experiment was focused on the area of manufacturing technology – milling. Have been investigated effects of different cutting conditions on the surface temperature of the machined sample and the results are expressed mainly in the charts.

This work deals with the measurement of temperature fields in the air using an infrared camera. The dissertation describes the opportunity of measuring the temperature field in the air and the characterization of the developed measuring method. In the next part there are introduced the beginning of thermography imaging and the field of usability of the new infrared measuring method. Further, the theoretical foundations of the thermography measuring method in the temperature fields in the air are described. In the theoretic background there are described the analysis of heat conduction in an auxiliary material, the determination of dynamic properties of the method and the analysis of radiation, which has an influence on infrared imagining. This method requires an insertion of the auxiliary material into the non-isothermal air flow, which can allow for the study of the temperature distribution in air. For effective visualization of temperature fields in the air using an infrared camera, the selection of the appropriate auxiliary material, on which the air temperature displays, is crucial. In the next part of the doctoral thesis, there is a description of static measuring properties of auxiliary materials. The usability range of the measuring method is determined from these properties. In the thesis there are presented the description of the device for the measurement of 2D temperature fields in the air and the description of the measuring device for 3D measurements of temperature fields in the air using an infrared camera, which can also be used for measurements of temperature fields in small enclosed spaces through a viewing window. For the practical use of the method, the detailed methodology of measuring temperature fields in the air by an infrared camera was developed and its applicability was demonstrated on practice examples. The developed measuring method can be used in many areas of research and in practice.

Mini Unmanned Aerial Vehicles (mUAVs) have the potential to be a great asset to Wilderness Search and Rescue groups by providing a bird's eye view of the search area. These vehicles can carry a variety of sensors to better understand the world below. This paper proposes using both Infrared (IR) and Visible Spectrum cameras on a mUAV for Wilderness Search and Rescue. It details a method for combining the color and heat information from these two cameras into a single fused display to reduce needed screen space for remote field use. To align the video frames for fusion, a method for simultaneously pre-calibrating the intrinsic and extrinsic parameters of the cameras and their mount using a single multi-spectral calibration rig is also presented. A user study conducted to validate the proposed image fusion methods showed no reduction in performance when detecting objects of interest in the single-screen fused display compared to a side-by-side display. Furthermore, the users' increased performance on a simultaneous auditory task showed that increased performance on a simultaneous auditory task showed that their cognitive load was reduced when using the fused display.

Le mouvement d’humidité dans le réseau poreux de certains matériaux est très souvent à l’origine de phénomènes préjudiciables pour la durabilité des constructions du génie civil. C’est en particulier le cas pour le séchage du bois, générateur de fissures et de délaminations aux interfaces de collage, et pour le béton en situation d’incendie où le mouvement d’humidité peut induire des désordres irréversibles (écaillage). Le recours à des modèles prédictifs de ruine des structures nécessite donc la simulation du mouvement d’humidité au sein des matériaux. Ces modèles de transfert de masse et de chaleur sont sophistiqués et nécessitent d’être confrontés à des mesures afin d’être validés. Peu de techniques expérimentales existent pour mesurer les mouvements ou gradients d’humidité dans les réseaux poreux, en particulier en régime transitoire (séchage, incendie). Les techniques existantes sont de plus généralement coûteuses et imposent des conditions sévères de sécurité pour les chercheurs. L’objectif de la thèse est donc de mettre au point un dispositif de mesure de gradients d’humidité basé sur l’imagerie Térahertz. Il s’agit d’une technique de mesure relativement peu onéreuse et permettant de réaliser des mesures en régime transitoire. Un banc expérimental existant sera donc adapté à la mesure du champ d’humidité sur éprouvettes de bois en conditions thermo-hydriques variables, et sur éprouvettes de béton en situation de chauffage. Les résultats constitueront une base de données utile à la compréhension des phénomènes de dégradation des matériaux et seront directement utilisables comme outil de validation de modèles de calcul. / The movement of moisture in the porous network of certain materials is very often at the origin of phenomena prejudicial to the durability of the constructions of the civil engineering. This is particularly the case for the drying of wood, which creates cracks and delaminations at bonding interfaces, and for concrete in situations of fire where the movement of moisture can induce irreversible disorders (chipping). The use of predictive models of structural ruin therefore requires the simulation of the moisture movement within the materials. These mass and heat transfer models are sophisticated and need to be confronted with measurements in order to be validated. Few experimental techniques exist to measure moisture movements or gradients in porous networks, especially in transient conditions (drying, fire). Existing techniques are often expensive and impose severe conditions of safety for the researchers. The objective of the thesis is therefore to develop a device for measuring gradients of moisture based on Terahertz imagery. This is a comparatively inexpensive measuring technique and makes it possible to carry out transient measurements. An existing experimental bench will therefore be adapted to the measurement of the moisture field on wood specimens under variable water-moisture conditions and on concrete specimens in a heating situation. The results will constitute a database useful for understanding the phenomena of degradation of materials and will be directly usable as a validation tool for calculation models.

This thesis investigates the application of index based, region segmentation, and deep learning methods to the sensor fusion of omnidirectional (O-D) Infrared (IR) sensors, Kinnect sensors, and O-D vision sensors to increase the level of intelligent perception for unmanned robotic platforms. The goals of this work is first to provide a more robust calibration approach and improve the calibration of low resolution and noisy IR O-D cameras. Then our goal was to explore the best approach to sensor fusion for vegetation detection. We looked at index based, region segmentation, and deep learning methods and compared them with a goal of significant reduction in false positives while maintaining reasonable vegetation detection. The results are as follows: Direct Spherical Calibration of the IR camera provided a more consistent and robust calibration board capture and resulted in the best overall calibration results with sub-pixel accuracy The best approach for sensor fusion for vegetation detection was the deep learning approach, the three methods are detailed in the following chapters with the results summarized here. Modified Normalized Difference Vegetation Index approach achieved 86.74% recognition and 32.5% false positive, with peaks to 80% Thermal Region Fusion (TRF) achieved a lower recognition rate at 75.16% but reduced false positives to 11.75% (a 64% reduction) Our Deep Learning Fusion Network (DeepFuseNet) results demonstrated that deep learning approach showed the best results with a significant (92%) reduction in false positives when compared to our modified normalized difference vegetation index approach. The recognition was 95.6% with 2% false positive. Current approaches are primarily focused on O-D color vision for localization, mapping, and tracking and do not adequately address the application of these sensors to vegetation detection. We will demonstrate the contradiction between current approaches and our deep sensor fusion (DeepFuseNet) for vegetation detection. The combination of O-D IR and O-D color vision coupled with deep learning for the extraction of vegetation material type, has great potential for robot perception. This thesis will look at two architectures: 1) the application of Autoencoders Feature Extractors feeding a deep Convolution Neural Network (CNN) fusion network (DeepFuseNet), and 2) Bottleneck CNN feature extractors feeding a deep CNN fusion network (DeepFuseNet) for the fusion of O-D IR and O-D visual sensors. We show that the vegetation recognition rate and the number of false detects inherent in the classical indices based spectral decomposition are greatly improved using our DeepFuseNet architecture. We first investigate the calibration of omnidirectional infrared (IR) camera for intelligent perception applications. The low resolution omnidirectional (O-D) IR image edge boundaries are not as sharp as with color vision cameras, and as a result, the standard calibration methods were harder to use and less accurate with the low definition of the omnidirectional IR camera. In order to more fully address omnidirectional IR camera calibration, we propose a new calibration grid center coordinates control point discovery methodology and a Direct Spherical Calibration (DSC) approach for a more robust and accurate method of calibration. DSC addresses the limitations of the existing methods by using the spherical coordinates of the centroid of the calibration board to directly triangulate the location of the camera center and iteratively solve for the camera parameters. We compare DSC to three Baseline visual calibration methodologies and augment them with additional output of the spherical results for comparison. We also look at the optimum number of calibration boards using an evolutionary algorithm and Pareto optimization to find the best method and combination of accuracy, methodology and number of calibration boards. The benefits of DSC are more efficient calibration board geometry selection, and better accuracy than the three Baseline visual calibration methodologies. In the context of vegetation detection, the fusion of omnidirectional (O-D) Infrared (IR) and color vision sensors may increase the level of vegetation perception for unmanned robotic platforms. A literature search found no significant research in our area of interest. The fusion of O-D IR and O-D color vision sensors for the extraction of feature material type has not been adequately addressed. We will look at augmenting indices based spectral decomposition with IR region based spectral decomposition to address the number of false detects inherent in indices based spectral decomposition alone. Our work shows that the fusion of the Normalized Difference Vegetation Index (NDVI) from the O-D color camera fused with the IR thresholded signature region associated with the vegetation region, minimizes the number of false detects seen with NDVI alone. The contribution of this work is the demonstration of two new techniques, Thresholded Region Fusion (TRF) technique for the fusion of O-D IR and O-D Color. We also look at the Kinect vision sensor fused with the O-D IR camera. Our experimental validation demonstrates a 64% reduction in false detects in our method compared to classical indices based detection. We finally compare our DeepFuseNet results with our previous work with Normalized Difference Vegetation index (NDVI) and IR region based spectral fusion. This current work shows that the fusion of the O-D IR and O-D visual streams utilizing our DeepFuseNet deep learning approach out performs the previous NVDI fused with far infrared region segmentation. Our experimental validation demonstrates an 92% reduction in false detects in our method compared to classical indices based detection. This work contributes a new technique for the fusion of O-D vision and O-D IR sensors using two deep CNN feature extractors feeding into a fully connected CNN Network (DeepFuseNet).

At present, the quality of image taken from infrared cameras is low compared to the other cameras because of manufacturing technology. So, resolution enhancement processes are becoming more important for these cameras. Super resolution is a good approach to solve this resolution problem. In general, the systems that infrared cameras used require video processing to perform in real time. So, a suitable approach should be selected and implemented to work in real time. The computational load and processing time are big issues in this case. FPGAs are proven to be suitable hardware devices for these types of works. Super resolution involves two parts as global motion estimation and high resolution image reconstruction. In this study, one suitable algorithm, namely as PM, for global motion estimation in the literature is selected to be implemented in real time. On the other hand, for high resolution image reconstruction part, FPGA structures of some well known algorithms in the literature, namely as POCS, MLE, MAP and LMS are proposed and their performance, resource requirements and timing considerations are discussed. Most efficient one is selected and implemented in FPGA.