This thesis explains the development of a computer vision system able to find and orient relatively small objects. The motivations is exchanging a monotonous work done by hand and replace it with an automation system with help of an ABB IRB 140 industrial robot. The vision system runs on a standard PC and is developed using the OpenCV environment, originally made by Intel in Russia. The algorithms of the system is written in C++ and the user interface in C++/CLI. With a derived test case, multiple vision algorithms is tested and evaluated for this kind of application. The result shows that SIFT/SURF works poorly with multiple instances of the search object and HAAR classifiers produces many false positives. Template matching with image moment calculation gave a satisfying result regarding multiple object in the scene and produces no false positives. Drawbacks of the selected algorithm developed where sensibility to light invariance and lack of performance in a skewed scene. The report also contains suggestions on how to precede with further improvements or research.

Many modern mobile applications, such as Unmanned Aerial Vehicles (UAVs), require sophisticated processing capability with low power consumption in a small form factor. UAVs, for example, may require a platform capable of controlling a camera, performing digital signal processing techniques on the pictures to detect faces or motion, and guiding the vehicle based on decisions made from the processed data. Additionally, since the vehicle is mobile and aerial, its effectiveness is heavily dependent on the size and power consumption of the platform. In this report, we explore this set of requirements and how well they are met with a Texas Instruments OMAP SoC on a BeagleBoard. Specifically, we report on the computational performance and power drawn by the OMAP General Purpose Processor (GPP) when performing a facial detection algorithm with OpenCV. We also analyze the performance enhancement possible by offloading the facial detection algorithm to the OMAP DSP coprocessor. In summary we find that the Beagleboard would be an appropriate platform for a simpler UAV capable of pre-processing still images taken every few seconds, but not for processing video data real-time. We conclude by describing other applications that are suitable for the Beagleboard. / text

Abstrakt Röntgen är ett område där man ställer frågan om processen skulle kunna automatiseras för att göra den enklare för sjuksköterskor. På så sätt ökar antalet patienter som kan röntgas eftersom det skulle gå snabbare. Med hjälp av datorseende och en servostyrd röntgenkamera kan man förverkliga delar av dessa drömmar genom att låta röntgenkameran själv justera sig efter en patient och även flyttas till en vald kroppsdel. Här undersöks och testas open-source biblioteket OpenCV. En prototyp på ett automatiskt system tas fram med syftet att testa OpenCVs funktionalitet och besvara ett antal frågor: Hur kan röntgen automatiseras genom användning av open-source programvarubibliotek med inriktning på bildbehandling? Vilka för – och nackdelar kan användandet av ett datorseendebibliotek vara? Kan man med dagens teknik utveckla en automatisk lösning som kan göras till en kommersiell produkt?

No description available.

No description available.

No description available.

No description available.

Computer vision and image processing systems are ubiquitous in automotive domain and manufacturing industry. Lane detection warning systems has been an elementary part of the modern automotive industry. Due to the recent progress in the computer vision and image processing methods, economical and flexible use of computer vision is now pervasive and computing with images is not just for the realm of the science, but also for the arts and social science and even for hobbyists. Image processing is a key technology in automotive industry, even now there is hardly a single manufacturing process that is thinkable without imaging. The applications of image processing and computer vision methods in embedded systems platform, is an ongoing research area since many years. OpenCV, an open-source computer vision library containing optimized algorithms and methods for designing and implementing applications based on video and image processing techniques. These method are organized in the form of modules for specific field including, user-graphic interface, machine learning, feature extraction etc [43]. Vision-based automotive application systems become an important mechanism for lane detection and warning systems to alert a driver about the road in localization of the vehicle [1]. In automotive electronic market, for lane detection problem, vision-based approaches has been designed and developed using different electronic hardware and software components including wireless sensor, camera module, Field-Programmable Gate Array (FPGA) based systems, GPU and digital signal processors (DSP) [13]. The software module consists on the top of real-time operating systems and hardware description programming language including Verilog, or VHDL. One of the most time critical task of vision based systems is to test system applications in real physical environment with wide variety of driving scenarios and validating the whole systems as per the automotive industry standards. For validating and testing the advanced driver assistance systems, there are some commercial tools available including Assist ADTF from Elektrobit, EB company [43]. In addition to the design and strict real-time requirements for advanced driver assistance systems applications based on electronic components and embedded platform, the complexity and characteristics of the implemented algorithms are two parameters that need to be taken into consideration choosing hardware and software component [13]. The development of vision-based automotive application, based on alone electronic and micro-controller is not a feasible solution approach [35] [13] and GPU based solution are attractive but has many other issues including power consumption. In this thesis project, image and video processing module is used from OpenCV library for road lane detection problems. In proposed lane detection methods, low-level image processing algorithms and methods are used to extract relevant information for lane detection problem by applying contrast analysis at pixel level intensity values. Furthermore, the work at hand presents different approaches for solving relevant partial problems in the domain of lane detection. The aim of the work is to apply contrast analysis based on low-level image processing methods to extract relevant lane model information from the grid of intensity values of pixel elements available in image frame. The approaches presented in this project work are based on contrast analysis of binary mask image frame extracted after applying range threshold. A set of points, available in an image frame, based lane feature models are used for detecting lanes on color image frame captured from video. For the performance measurement and evaluation, the proposed methods are tested on different systems setup, including Linux, Microsoft Windows, CodeBlocks, Visual Studio 2012 and Linux based Rasbian-Jessie operating systems running on Intel i3, AMD A8 APU, and embedded systems based (Raspberry Pi 2 Model B) ARM v7 processor respectively.

The production line of industries are getting more efficient and having very high throughput. Different kinds of machineries are being used to make the production safe, fast, precise and reliable. Robot arm is such a machine which helps the production line to be more efficient and productive. Nowadays, many manufacturing industries are using robot-arms to get a competitive edge in manufacturing and can be outfitted for multiple applications like welding, material handling, thermal spraying, painting, drilling and so on. They are widely used to increase product quality and production demand and over all, to ensure safer, faster and efficient production. It is very important to control and maintain these machines very accurately. As a simple mistake of robot arm can cause excessive destructions and bring financial losses to the industries, the robotarms must be very accurate when they are functioning in their production settings.

The purpose of this thesis is to develop the architecture to use the Car2X for observation the local traffic sign and displays it on the OpenStreetMap to provide more information of the road side to the driver. The proposed architecture of this thesis is to convert the traffic sign into the barcode and to be scanned by the barcode scanner and then wirelessly transfers the data to the web server to store the data and displays the traffic sign on the OpenStreetMap in the web browser. It uses two Raspberry Pi boards with CAN-Bus shields for transmitting the data on the CAN-Bus system in the car, a barcode scanner to scan the barcode, a GPS module to get its location, and a WiFi dongle to wirelessly send the data. The thesis also includes the camera to detect the traffic light using OpenCV and sends the GO or STOP command to the car. The results provide the OpenStreetMap with the traffic sign which helps the driver to realize the traffic sign on the road of the desired destination. However, the accuracy of GPS is not satisfied as well as the distance of the barcode scanning, therefore, this thesis suggests that includes the gps position in the barcode and uses the camera to detect the barcode for the improvement in the future.