Browsing by Author "Ekanayake, Thushara"

Now showing 1 - 2 of 2
  • No Thumbnail Available
    Item
    Convert 2D Images to 3D Image
    (Uva Wellassa University of Sri Lanka, 2012) Menikdiwela, M. P.; Ekanayake, Thushara
    Image processing gives a major contribution to the modern technology such as robot navigation, medicine, video games, three dimensional televisions etc. And also there are some places we cannot ever reach those places, which are war areas, space, deep sea etc. But we can get satellite images. Those images can be processed by image processing techniques. Therefore, recently image processing is a widely developing part of the technology. The 3D view means third dimension of depth which can be perceived by the human vision in the form of binocular inequality. Human eyes are located at slightly different positions which can observe different views of the real world. The brain is then able to reconstruct the depth information from these different views. A 3D display takes advantage of this phenomenon, creating two slightly different images of every scene and then displaying them to the individual eyes. With an appropriate disparity and calibration of parameters, a correct 3D perception can be identified 2D to 3D conversion is also an important step in image processing. By the support of MatlabOD software, it is possible to develop algorithms for 2D to 3D image conversion. MatlabĀ® is one of the major software in image processing, because it has its own programming language. Whole algorithm is run by the concept of stereovision. With two images of the same scene captured from the slightly different views and get the 3D view is identified as stereovision. Binocular stereo vision uses only two images.
  • No Thumbnail Available
    Item
    Outdoor Robotic Localization
    (Uva Wellassa University of Sri Lanka, 2012) Watawana, B.J.; Piyathilaka, Lasitha; Ekanayake, Thushara
    Robotic Localization is a vast area of research & implemented in different scales. Many researches are carried throughout the world to find optimum and most cost effective method of implementing a localization system. In these researches, technologies like Computer Vision, GPS, Wi-Fi, GSM, RFID and ultrasonic are used for the localization. The main concept of robotic localization is similar where it uses a known stationary position as the base and calculates the relative position with different technologies. Outdoor localization involves handling environmental ambient noises as well as adapting system to sudden changes in parameters such as light level variation. The motivation of this research is to address this issue and find an optimum trilateration algorithm for robotic localization. CV, Wi-Fi, GSM, RFID and Ultrasonic sensory localization uses trilateration algorithms. Trilateration algorithms are used for localizing robot in a limited area with a higher accuracy. Researchers are carried out in trilateration algorithm either using ultrasonic sensors or computer vision. Ultrasonic sensory trilateration involves higher error ratio due to variations of environmental air velocity, moisture level and air density. CV mainly lacks adjustability to light condition variations. Higher the accuracy in CV, greater the cost of implementation. This research is based upon combining both algorithm to get a higher accuracy output and making the algorithm viable with higher ambient noise with a low cost