An Anomaly Detection System For Transparent Objects Using Polarizedimage Fusion Technique

• doi: 10.1109/SAS54819.2022.9881251
• title: An Anomaly Detection System for Transparent Objects Using Polarized-Image Fusion Technique
• publisher: IEEE
• isbn: 978-1-6654-0982-7
• issn:
• rank: 1476
• access_type: LOCKED
• content_type: Conferences
• abstract: An anomaly detection system using a polarized-image fusion technique has been developed for food inspection applications. It is capable of detecting (a) foreign objects among objects wrapped in transparent reflective material and (b) transparent foreign objects in transparent bottles. The conventional anomaly detection system using a traditional RGB camera has low accuracy for such detection, due to the large amount of glare that can occur from reflective surfaces. Regions with glare are often falsely perceived as anomalies. Since transparent foreign objects have few features, they are difficult to recognize. To address these problems, a polarized-image fusion (PIF) technique is developed. Four polarized images are fused to synthesize a high-quality image where glare is suppressed, and transparent foreign objects are highlighted. These polarized images are captured simultaneously by a single camera utilizing an advanced polarized CMOS image sensor. The PIF technique was evaluated with two kinds of data set: (1) cookie samples wrapped in transparent plastic bags and (2) transparent plastic bottles containing transparent plastic foreign objects. High anomaly detection accuracies of 0.851 AUC (area under receiver operating characteristic curve) for the cookie sample data set and 0.871 AUC for the plastic bottle data set were achieved. Compared with the deep one-class classification neural network with simple RGB data input, the accuracies were improved by 0.09 AUC for both cases.
• article_number: 9881251
• pdf_url: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=9881251
• html_url: https://ieeexplore.ieee.org/document/9881251/
• abstract_url: https://ieeexplore.ieee.org/document/9881251/
• publication_title: 2022 IEEE Sensors Applications Symposium (SAS)
• conference_location: Sundsvall, Sweden
• conference_dates: 1-3 Aug. 2022
• publication_number: 9881325
• is_number: 9881241
• publication_year: 2022
• publication_date: 1-3 Aug. 2022
• start_page: 1
• end_page: 6
• citing_paper_count: 5
• citing_patent_count: 0
• download_count: 278
• insert_date: 20220912
• index_terms:
  • ieee_terms:
    • Signal processing algorithms
    • Signal processing
    • Sensor fusion
    • Cameras
    • Sensors
    • Plastic products
    • Plastics
  • author_terms:
    • Anomaly detection
    • Deep learning
    • Polarization
    • Image fusion
  • dynamic_index_terms:
    • Anomaly Detection
    • Fusion Techniques
    • Transparent Objects
    • Anomaly Detection Systems
    • Neural Network
    • Deep Neural Network
    • Plastic Bags
    • Foreign Body
    • Foreign Objects
    • High-quality Images
    • High Quality Images
    • Image Sensor
    • Camera Sensor
    • Surface Reflectance
    • Material Objects
    • Inanimate Objects
    • Plastic Bottles
    • Transparent Plastic
    • Kinds Of Datasets
    • Kinds Of Data Sets
    • Polarization Imaging
    • One-class Classification
    • Exposure Time
    • Support Vector Machine
    • Dynamic Range
    • Strong Reflection
    • Specular Reflection
    • Polar Angle
    • Brewster Angle
    • Automated Guided Vehicles
    • Autonomous Guided Vehicles
    • Autonomous Mobile Robots
    • Object Position
    • Soft Objects
    • Weight Map
    • High Dynamic Range
    • High-dynamic-range
    • Input Image
    • Fusion Algorithm
• isbn_formats:
  • format: Print on Demand(PoD) ISBN, value: 978-1-6654-0982-7, isbnType: New-2005
  • format: Electronic ISBN, value: 978-1-6654-0981-0, isbnType: New-2005
• authors:
  • Author Name: Lixing Yu
    Affiliation: Dept. of Electrical Engineering and Information Systems, The University of Tokyo, Tokyo, Japan
    Author URL: https://ieeexplore.ieee.org/author/37089524331
    ID: 37089524331
    Order: 1
    Author Affiliations:
    • Dept. of Electrical Engineering and Information Systems, The University of Tokyo, Tokyo, Japan
  • Author Name: Atsutake Kosuge
    Affiliation: Systems Design Laboratory, The University of Tokyo, Tokyo, Japan
    Author URL: https://ieeexplore.ieee.org/author/38242741200
    ID: 38242741200
    Order: 2
    Author Affiliations:
    • Systems Design Laboratory, The University of Tokyo, Tokyo, Japan
  • Author Name: Mototsugu Hamada
    Affiliation: Systems Design Laboratory, The University of Tokyo, Tokyo, Japan
    Author URL: https://ieeexplore.ieee.org/author/37286985900
    ID: 37286985900
    Order: 3
    Author Affiliations:
    • Systems Design Laboratory, The University of Tokyo, Tokyo, Japan
  • Author Name: Tadahiro Kuroda
    Affiliation: Systems Design Laboratory, The University of Tokyo, Tokyo, Japan
    Author URL: https://ieeexplore.ieee.org/author/37274599600
    ID: 37274599600
    Order: 4
    Author Affiliations:
    • Systems Design Laboratory, The University of Tokyo, Tokyo, Japan

Image Sensor

• sensor_type: CMOS
• resolution: 1224x1024
• dynamic_range: Not specified
• pixel_size: Not specified
• dark_current: Not specified

Optical Data

• focal_length: Not specified
• aperture: Not specified
• field_of_view: Not specified
• distortion: Not specified

Performance Metrics

Applications & Benefits

• cell_imaging: Enabled by image sensors for applications like food inspection and anomaly detection.
• benefits: Improved accuracy in detecting transparent foreign objects and reflective surfaces.

Supporting Organizations

• supported_by: New Energy and Industrial Technology Development Organization (NEDO)

Manuscript Details

• publication_date: 1-3 Aug. 2022

Relevancy Score

• score: 9
• missing_fields:
  • dynamic_range
  • fill factor
  • quantum efficiency
  • dark_current
  • signal-to-noise ratio
  • power consumption
  • focal_length
  • aperture
  • field_of_view
  • distortion
  • frame_rate
  • sensitivity
  • shutter_speed
  • noise

Processed JSON Filename

request_0b62d5ba-1065-4865-8f00-e1ba7842ebf5-an_anomaly_detection_system_for_transparent_objects_using_polarizedimage_fusion_technique.json