Automatic Fog Detection in Day and Night Images to Improve Highway Driving Conditions

• Other Authors: Patel, Jigar Kamlesh (authoraut)
• Format: Others
• Language: English; English
• Published: Florida State University
• Subjects: Electrical engineering
• Online Access: http://purl.flvc.org/fsu/fd/FSU_FA2016_Patel_fsu_0071N_13465

Visibility is an important part of highway driving. Weather-related visibility limitations have caused several major pile-ups (crashes) over the years all over the country. These limitations are caused due to, but not limited to, fog and rain. It is vital to be aware of the driving environment and conditions when traveling on interstates and highways at speeds upwards of 70 mph. Crashes caused at these speeds usually result in injuries and sometimes death. In an effort to minimize/reduce the rate of such incidents, this project intends on automatically detecting driving visibility ranges in real time based on camera systems that already exist on the Florida Highway System. This will be done to alert travelers of driving hazards due to visibility limitations. We plan on doing this by performing analysis on images taken over time to detect the density of fog to determine the visibility distance. This thesis presents a conceptual development, theoretical design, implementation, and the implementation results to achieve automatic fog detection in day-time and night-time images on a standard scale regardless of the image intensity. (i.e. Day-time and Night-time images have the same scale). Images are considered independently with no reference image to compare with. A review of related work is conducted for fog detection and sharpness measure of images to achieve this goal. === A Thesis submitted to the Department of Computer & Electrical Engineering in partial fulfillment of the requirements for the degree of Master of Science. === Fall Semester 2016. === September 09, 2016. === Fog detection === Includes bibliographical references. === Victor DeBrunner, Professor Directing Thesis; Uwe Meyer-Baese, Committee Member; Bruce Harvey, Committee Member.