Advanced

Ionizing Radiation Effects on Image Sensors: Method on Evaluation from an Image Quality Perspective

Fagerström, Johan (2016)
Computer Science and Engineering (BSc)
Abstract
Surveillance cameras are vital in the nuclear power industry where the reactor equipment must constantly be kept under surveillance to maintain safety. Constant monitoring is however difficult due to the highly radioactive environment, which cause cameras to malfunction and eventually break. Malfunction typically manifests itself as quality degradation of the image from the cameras. Initially, temporal noise appears on the image, while after some time in operation the image can be so distorted that the camera no longer serves its purpose. The image sensor, which exists in every electronic camera, is a key component as it converts the physical image (visible light) to an electronic image. As ionizing radiation and light have similar... (More)
Surveillance cameras are vital in the nuclear power industry where the reactor equipment must constantly be kept under surveillance to maintain safety. Constant monitoring is however difficult due to the highly radioactive environment, which cause cameras to malfunction and eventually break. Malfunction typically manifests itself as quality degradation of the image from the cameras. Initially, temporal noise appears on the image, while after some time in operation the image can be so distorted that the camera no longer serves its purpose. The image sensor, which exists in every electronic camera, is a key component as it converts the physical image (visible light) to an electronic image. As ionizing radiation and light have similar properties, image sensors are also sensitive to ionizing radiation. Previous work has evaluated various physical effects that ionizing radiation has on image sensors. Common is that the effects are visible on the image sensors output and imply a substantial impact on image quality. However, the main problem with previous evaluations is the difficulty to value and compare the results in respect to image quality impact. For example, it is difficult to determine whether glass browning causes more quality impact than dark current increase. To address these problems, in this thesis we propose a method that evaluates the effects purely from the perspective of image quality. Metrics developed in the area of image quality assessment are used for measuring the image quality degradation. The radiation impact on image sensors is thus evaluated and quantified in a way that is more comprehensive, and where the results are easier valued and easier compared. Using the proposed method, questions such as "Are CCD sensors more radiation tolerant than CMOS sensors?" can be answered in a new sense. The method is demonstrated upon four Omnivision OV7949 image sensors. The sensors are irradiated up to 1600 Gy in a TRIGA Mark II nuclear research reactor while gathering raw data. After finished irradiation process, the data is analyzed and the quality degradation induced by ionizing radiation is assessed and expressed in relation to absorbed dose. The result is visualized as plotted graphs, and three measurements are made: initial-, cumulative and total degradation. To show how the results can be utilized and how the results are easily compared, we make an example of determining the best performing image sensor among the four evaluated. (Less)
Please use this url to cite or link to this publication:
author
Fagerström, Johan
organization
year
type
M2 - Bachelor Degree
subject
keywords
image sensor, cmos, ov7949, electronic camera, radiation effects, ionizing radiation, image quality assessment, radiation tolerance
language
English
id
8877372
alternative location
https://portal.ch.lu.se/Campus.NET/Services/Publication/Export.aspx?id=2921&type=doc
date added to LUP
2016-06-04 04:07:42
date last changed
2016-06-04 04:07:42
@misc{8877372,
  abstract     = {Surveillance cameras are vital in the nuclear power industry where the reactor equipment must constantly be kept under surveillance to maintain safety. Constant monitoring is however difficult due to the highly radioactive environment, which cause cameras to malfunction and eventually break. Malfunction typically manifests itself as quality degradation of the image from the cameras. Initially, temporal noise appears on the image, while after some time in operation the image can be so distorted that the camera no longer serves its purpose. The image sensor, which exists in every electronic camera, is a key component as it converts the physical image (visible light) to an electronic image. As ionizing radiation and light have similar properties, image sensors are also sensitive to ionizing radiation. Previous work has evaluated various physical effects that ionizing radiation has on image sensors. Common is that the effects are visible on the image sensors output and imply a substantial impact on image quality. However, the main problem with previous evaluations is the difficulty to value and compare the results in respect to image quality impact. For example, it is difficult to determine whether glass browning causes more quality impact than dark current increase. To address these problems, in this thesis we propose a method that evaluates the effects purely from the perspective of image quality. Metrics developed in the area of image quality assessment are used for measuring the image quality degradation. The radiation impact on image sensors is thus evaluated and quantified in a way that is more comprehensive, and where the results are easier valued and easier compared. Using the proposed method, questions such as "Are CCD sensors more radiation tolerant than CMOS sensors?" can be answered in a new sense. The method is demonstrated upon four Omnivision OV7949 image sensors. The sensors are irradiated up to 1600 Gy in a TRIGA Mark II nuclear research reactor while gathering raw data. After finished irradiation process, the data is analyzed and the quality degradation induced by ionizing radiation is assessed and expressed in relation to absorbed dose. The result is visualized as plotted graphs, and three measurements are made: initial-, cumulative and total degradation. To show how the results can be utilized and how the results are easily compared, we make an example of determining the best performing image sensor among the four evaluated.},
  author       = {Fagerström, Johan},
  keyword      = {image sensor,cmos,ov7949,electronic camera,radiation effects,ionizing radiation,image quality assessment,radiation tolerance},
  language     = {eng},
  note         = {Student Paper},
  title        = {Ionizing Radiation Effects on Image Sensors: Method on Evaluation from an Image Quality Perspective},
  year         = {2016},
}