Med en ökad efterfrågan av övervakningskameror på marknaden ökar samtidigt efterfrågan för olika monteringsmöjligheter. I andra länder, utanför Sverige, användskameror för att filma vägen inifrån fordonen, vilket nu håller på att introduceras iSverige. En sådan kamera sitter exempelvis på instrumentbrädan eller på vindrutanoch kallas dashcam. Detta är en ny marknad inom Sverige och omfattar trafikövervakning via fordon. Detta examensarbete utförs på Malmö universitet i samarbetemed Axis Communications AB. Syftet med detta examensarbete är att utforma enfast montering av en kamera på insidan av vindrutan i fordon för att fungera som endashcam, genom att använda olika processer av produktutveckling och andra metoder. Målet är att utveckla en prototyp som ska fungera som ett underlag för framtidautveckling. Konceptet som valdes i detta projekt fokuserar på en helhetslösning föratt kunna filma ut ur fordon med acceptabel bildkvalité. Detta resulterade i en prototyp som monteras på insidan av ett fordon mot vindrutan med en trattlösning föratt minimera reflektioner från oönskat ljus för att åstadkomma en bra bild. Genomavgränsningar i projektet har en bredare helhetslösning utvecklats.Syftet och målet för projektet anses vara uppfyllda då de flesta krav enligt önskadekundbehov har blivit uppfyllda och prototypen är anpassad för de flest önskvärdafunktioner. Arbetet har fokuserat på att kommentera punkter värt att undersöka förframtida utveckling. / With an increased demand on the market for surveillance cameras, there is a increaseddemand at the same time for different mounting possibilities. In other countries, outside of Sweden, the usage of cameras has been applied to record the road from insideof the vehicle, this is now presented in Sweden. Said camera is usually installed onthe dashboard or windshield and is called a dashcam. This is a new market withinSweden and involves traffic surveillance by vehicle. This thesis is performed at MalmöUniversity in collaboration with Axis Communications AB. The purpose of this thesisis to develop a solid mounting of a camera on the inside of a windshield of a vehicle to work as a dashcam, by using different processes of product development andother methods. The goal is to develop a prototype to be used as a basis for futuredevelopments. The concept chosen in this project focus on the total solution for beingable to record from the inside of the vehicle with an acceptable recording quality.This resulted in a prototype which is mounted on the inside of the vehicle againstthe windshield with a funnel solution to minimize the reflections from unwanted lightto achieve a good recording quality. By delimitations of the project a broader totalsolution could be developed.The purpose and goal for this project is considered fulfilled. As most requirementsaccording to the desired customer needs is fulfilled and the prototype is adapted tothe most desired functions. The thesis has also focused on commenting subjects worthinvestigating in future development.

Food security is a major concern of the 21st century, given climate change and population growth. In addition, high salt concentrations in soils affect ~20% of irrigated land and cause a substantial reduction in crop yield. Cultivating salt-tolerant crops could enable the use of salt-affected agricultural land, reduce the use of fresh water and alleviate yield losses. Innovative methods need to be developed to study traditional and novel traits that contribute to salinity tolerance and accurately quantify them. These studies would eventually serve for developing new salt tolerant crops, adapted to the harsh arid and semi-arid climate conditions. A study of 200 accessions of the wild tomatoes (Solanum pimpinellifolium) was conducted in field conditions with phenotyping using an unmanned aerial vehicle (UAV)-mounted hyperspectral camera. Six genotypes with different levels of salt tolerance were sampled for leaf pigment analyses, revealing a clear pattern for the high salt tolerant accession M007, where pigment content in the salt-treated plants significantly increased compared to their control counterparts only in harvesting campaigns 3 and 6, each performed two days after the first and second salt stress application events. Moreover, the light harvesting capacity was found to be better maintained under salt stress in the medium (M255) and highly salt tolerant (M007 and M061) accessions. Pigment quantitation data will contribute towards the groundtruthing of hyperspectral imaging for the development of remote sensing-based predictive pigment mapping methods. This work establishes a reliable quantification protocol for correlating pigment content with vegetation indices. Hence, pigment content captured by imaging techniques and validated using biochemical analysis would serve in developing a high-throughput method for pigment quantitation in the field using UAV-based hyperspectral imaging. This would serve as a tool for measuring pigment content in large number of genotypes in the field which would eventually lead to new salt-tolerant genes.

Emerging "Structure-from-Motion" (SfM) photogrammetry techniques encourage faster, cheaper, and more accessible field methods for accurately reconstructing 3D topography. The SfM method consists of collecting sets of overlapping images of the ground surface with a point and shoot camera, and reconstructing surface topography from the images with developed software programs. This research develops and implements a SfM image acquisition method and post-processing workflow as a supplemental technique to the traditional total-station method to aid in monitoring sandbar change in Marble and Grand Canyons along the Colorado River in Arizona. Due to permitting in Grand Canyon National Park, a 4.9 m pole-mounted camera platform was used in this research to mimic the ground perspective of an aerial platform. This research presents an improved understanding of how the low-angle, pole-mounted camera platform affects image acquisition and ultimately 3D reconstructions of the surface topography. Models of ground surfaces always contain some degree of elevation error, or uncertainty. As such, elevation error models are needed to distinguish whether observed changes to topographic features (in this case sandbars) are real or simply due to elevation error. There are many ways to quantify multiple sources of elevation uncertainty, but in this study the sources of elevation uncertainty were considered to vary across the surface and were characterized accordingly. Especially in river environments with complex surface topography (e.g. steep cut banks), and roughness (e.g. vegetation), quantifying the spatially variable elevation uncertainty of the surface representation is critical for interpreting actual changes in surface topography over repeat surveys. This research: used the sandbar images collected in Marble and Grand Canyons with the pole-mounted camera platform to generate SfM, topographic models; calculated spatially variable surface uncertainty derived from slope and roughness using multiple statistical analyses; built an error model that was calibrated based upon the statistical analyses of the spatially variable surface uncertainty; Key findings of this research are: Densely vegetated topography results in high amounts of elevation uncertainty, and without additional information of the surface underlying the vegetation, the SfM tool is less operational in these areas; Bare, exposed topography with low to high slopes that are not covered in black shadows result in lower surface uncertainty, and are areas where SfM is an operational tool for studies of surface change. Complementing existing topographic sampling methods with more efficient and cost- effective SfM approaches will contribute to the understanding of changing responses of the topographic features. In addition, the development and implementation of SfM and corresponding amounts of elevation uncertainty for monitoring geomorphic change will provide a methodological foundation for extending the approach to other geomorphic systems world- wide.

Indiana University-Purdue University Indianapolis (IUPUI) / Over the years, car driving experience has evolved drastically. Many new and useful technologies have emerged, which have enhanced safety and reliability measures. The Automotive world is now trying to build capabilities for driverless or vehicle assisted driving. Building capabilities for driverless cars practically means first developing training methods, then training the machine, evaluating the test results, and then based on testing results; develop a confidence interval for trusting the machine. One of the critical models is the model adopting the Road Departure Assisting Techniques (RDAT). These techniques are primarily the standards for alleviating the risk of roadside fatalities. The different models developed or proposed for RDAT falls under “The Road Departure Mitigation System” (RDMS). But, almost every RDMS to date has over-reliance on the presence and the quality of the lane markings. In the absence of lane markings or of proper lane markings, these RDMS are unreliable. Therefore, RDMS requires new references such as roadside objects and road edges for detecting road departures. This new system should propose and establish a standard for RDMS testing with roadside objects. As the foremost task, this new system requires the creation of a testing environment consisting of soft, robust, and reusable surrogates. Critically, these surrogates must have comparable sensors characteristics to those of real roadside objects from various commonly used object detection sensors on the vehicles such as camera, radar, and LIDAR. One of such everyday roadside objects is the curbs. For developing a surrogate for the curb, the first step is to recognize what the roadside objects should look like concerning different sensors, and the next step is to design and develop a surrogate curb that successfully follows the properties of the real roadside objects. This thesis first demonstrates and proposes the methods for extracting the color, Radar reflectivity, and the LiDAR reflectance properties of real roadside curbs. That is, the study deals with what all color combinations and patterns represent the US roadside curbs, what should be the range of Radar reflectivity values, and LiDAR reflectance bounds that a surrogate curb should satisfy. The later part of the thesis illustrates methods and steps on how to mimic the extracted properties, design a surrogate curb as per federal standards, and then develop a surrogate curb. Finally, the surrogate curbs were subjected to crash tests for testing their robustness.

京都大学 / 0048 / 新制・課程博士 / 博士(情報学) / 甲第17919号 / 情博第501号 / 新制||情||89(附属図書館) / 30739 / 京都大学大学院情報学研究科知能情報学専攻 / (主査)教授 松山 隆司, 教授 美濃 導彦, 教授 中村 裕一 / 学位規則第4条第1項該当 / Doctor of Informatics / Kyoto University / DFAM

京都大学 / 0048 / 新制・課程博士 / 博士(理学) / 甲第20786号 / 理博第4330号 / 新制||理||1622(附属図書館) / 京都大学大学院理学研究科物理学・宇宙物理学専攻 / (主査)教授 谷森 達, 教授 永江 知文, 教授 鶴 剛 / 学位規則第4条第1項該当 / Doctor of Science / Kyoto University / DFAM

京都大学 / 0048 / 新制・課程博士 / 博士(情報学) / 甲第21910号 / 情博第693号 / 新制||情||119(附属図書館) / 京都大学大学院情報学研究科知能情報学専攻 / (主査)講師 延原 章平, 教授 西野 恒, 准教授 飯山 将晃 / 学位規則第4条第1項該当 / Doctor of Informatics / Kyoto University / DFAM

The aim of this thesis is to investigate how well a generic monocular camera, placed on the vehicle, can be employed to localize an autonomous vehicle in a warehouse setting. The main function is to ascertain which zone the vehicle is currently in, as well as update the status when entering a new zone. Two zones are defined, where one has a lower allowed top velocity and the other a higher one. For this purpose ArUco markers are used to signal the system as to where it currently is. Markers are strategically placed around the laboratory area to saturate the environment with possible detections. Multiple sequences are recorded while varying camera placement, angles, and paths to determine the optimal number and placement of markers. In addition to this, a SLAM solution is tested in order to explore what benefits can be found. The idea is to provide fine-grained localization as well as a map of the warehouse environment, to provide more options for further development. To solve the SLAM problem, an implemented particle filter approach initializes a set of particles uniformly distributed within the world frame. For each frame, the particles undergo pose prediction, weight assignment based on likelihood, and resampling. This iterative process gradually converges the particles toward the camera's true position. Visual odometry techniques are used to estimate the camera's ego-motion. The process involves acquiring a sequence of images, detecting distinctive features, matching features between consecutive frames, estimating camera motion, and optionally applying local optimization techniques for further refinement. The implementation shows promise and all test cases performed during the project have been successful as for the zone localization. The SLAM solution can detect and track specific features or landmarks over consecutive frames. By triangulating the positions of these features, their depth and distance can be determined. However, the visualization of these features on a top-down map, which was part of the plan, has not been completed yet despite finishing the particle filter implementation.

A new test set design and benchmarking approach (US Patent pending) allows a "standard observer" to assess the end-to-end image quality characteristics of video imaging systems operating in day time or low-light conditions. It uses randomized targets based on extensive application of Photometry, Geometrical Optics, and Digital Media. The benchmarking takes into account the target’s contrast sensitivity, its color characteristics, and several aspects of human vision such as visual acuity and dynamic response. The standard observer is part of the "extended video imaging system" (EVIS). The new test set allows image quality benchmarking by a panel of standard observers at the same time. The new approach shows that an unbiased assessment can be guaranteed. Manufacturers, system integrators, and end users will assess end-to-end performance by simulating a choice of different colors, luminance levels, and dynamic conditions in the laboratory or in permanent video systems installations.

The purpose of this research is to ascertain potential iris scan data variations from near infrared waves derived from fluorescent illumination. Prior studies of iris data variances from infrared wave interference of halogen, incandescent, and sunlight with iris cameras suggest that similar changes may exist under near infrared wavelengths from fluorescent light. The concern is that the fluorescent energy emission may interfere with the near infrared detection of an iris camera. An iris camera is used to measure human eye characteristics known as biometrics. If such infrared emission is statistically significant, then it can alter the validity of the iris scan data. The experiment utilized nine hundred forty-five (945) scans from sixty-three (63) subjects. Measured results showed increased heat from ambient fluorescent illumination does not statistically alter the biometric readings of human eyes. The test results fail to reject that data loss will not occur as heat is increased in the ambient fluorescent light source.