Automotive Integration Of Smart Road Condition Detection System

• doi: 10.1109/SSI58917.2023.10387757
• title: Automotive Integration of Smart Road Condition Detection System
• publisher: IEEE
• isbn: 979-8-3503-0231-8
• issn:
• rank: 1693
• access_type: LOCKED
• content_type: Conferences
• abstract: We introduce the integration of an innovative approach for optically assessing hazardous road conditions, such as detecting ice and wet surfaces. Our cutting-edge method effortlessly pairs with commercially available CMOS image sensors, expanding the use of standard optical filters and significantly boosting the safety features of selected Advanced Driver Assistance Systems (ADAS). The detection system has been designed with cost-effectiveness and high accuracy in mind, leveraging the physical optical properties of water in conjunction with information obtained from RGB camera recordings, rather than relying solely on the latter. The detection system has been designed to function without the need for additional stimulation, utilizing only naturally available light sources or artificial illumination such as headlights or streetlights. Rather than relying on absolute measurements, it evaluates road conditions through the utilization of spectral absorption contrast ratio and polarization contrast ratio. A simple CMOS sensor, devoid of the Bayer filter, captures the light scattered off the road surface in two spectral bands (VIS and NIR) and two polarization planes. A machine learning-based classification algorithm, utilizing a trained Support Vector Machine (SVM) model, is employed for data inference to classify the road condition. The integrated system has undergone initial field tests, demonstrating good selectivity and robustness in the presence of movement and changing ambient light conditions. This paper provides an overview of the system and details its integration into a conventional passenger vehicle, located behind the windshield window. The initial results and encountered challenges will be presented and discussed.
• article_number: 10387757
• pdf_url: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10387757
• html_url: https://ieeexplore.ieee.org/document/10387757/
• abstract_url: https://ieeexplore.ieee.org/document/10387757/
• publication_title: 2023 Smart Systems Integration Conference and Exhibition (SSI)
• conference_location: Brugge, Belgium
• conference_dates: 28-30 March 2023
• publication_number: 10387937
• is_number: 10387754
• publication_year: 2023
• publication_date: 28-30 March 2023
• start_page: 1
• end_page: 4
• citing_paper_count: 0
• citing_patent_count: 0
• download_count: 72
• insert_date: 20240117
• index_terms:
  • ieee_terms:
    • Optical filters
    • Support vector machines
    • Integrated optics
    • Water
    • Optical polarization
    • Roads
    • System integration
  • author_terms:
    • road condition detection
    • spectral absorption
    • NIR
    • polarization
  • dynamic_index_terms:
    • Detection System
    • Road Conditions
    • Support Vector Machine
    • Light Conditions
    • Spectral Bands
    • Linearly Polarized
    • Propagation Direction
    • Road Surface
    • Resurfacing
    • Unimproved
    • Surface Wettability
    • Image Sensor
    • Camera Sensor
    • Support Vector Machine Model
    • Standard Light
    • Street Lighting
    • System Overview
    • Spectral Content
    • Spectral Absorption
    • Windshield
    • Windscreen
    • Car Window
    • Advanced Driver Assistance Systems
    • Driver Assistance Systems
    • Passenger Vehicles
    • Presence Of Movements
    • Artificial Illumination
    • Visible Light
    • Visible Region
    • Visible Spectrum
    • Experimental System
    • Horizontal Polarization
    • Autonomous Vehicles
    • Autonomous Driving
    • Camera System
    • State Recognition
    • Recognition Of Status
    • Recognition Of Conditions
    • Vertical Polarization
    • Puddles
    • Perspective Camera
    • Detection Conditions
    • State Detection
• isbn_formats:
  • format: Print on Demand(PoD) ISBN, value: 979-8-3503-0231-8, isbnType: New-2005
  • format: Electronic ISBN, value: 979-8-3503-2506-5, isbnType: New-2005
• authors:
  • Author Name: Dietrich Dumler
    Affiliation: Machine Learning enhanced Sensor Systems, Fraunhofer Research Institution for Microsystems and Solid State Technologies, Muinich, Germany
    Author URL: https://ieeexplore.ieee.org/author/610475336939076
    ID: 610475336939076
    Order: 1
    Author Affiliations:
    • Machine Learning enhanced Sensor Systems, Fraunhofer Research Institution for Microsystems and Solid State Technologies, Muinich, Germany
  • Author Name: Franz Wenninger
    Affiliation: Machine Learning enhanced Sensor Systems, Fraunhofer Research Institution for Microsystems and Solid State Technologies, Muinich, Germany
    Author URL: https://ieeexplore.ieee.org/author/38124576300
    ID: 38124576300
    Order: 2
    Author Affiliations:
    • Machine Learning enhanced Sensor Systems, Fraunhofer Research Institution for Microsystems and Solid State Technologies, Muinich, Germany

Image Sensor

• sensor_type: CMOS
• resolution:
• dynamic_range:
• pixel_size:
• dark_current:

Optical Data

• focal_length:
• aperture:
• field_of_view:
• distortion:

Performance Metrics

• frame_rate:
• signal_to_noise_ratio:
• sensitivity:
• shutter_speed:
• power_consumption:
• noise:

Applications & Benefits

• cell_imaging:
• benefits:

Supporting Organizations

• supported_by:

Manuscript Details

• publication_date: 28-30 March 2023

Relevancy Score

• score: 10
• missing_fields:
  • resolution
  • dynamic_range
  • pixel_size
  • dark_current
  • focal_length
  • aperture
  • field_of_view
  • distortion
  • frame_rate
  • sensitivity
  • shutter_speed
  • power_consumption
  • noise

Processed JSON Filename

request_cd517cba-4b65-4730-b9a4-9b19e871b5a8-automotive_integration_of_smart_road_condition_detection_system.json