Introduction – Video – White papers – Documentation – Electronic Imaging 2024 – ISO 23654 – Call for participation
Introduction
The market for cameras that produce images for Machine vision (MV) and Artificial Intelligence (AI), in contrast to pictorial images for human vision, is steadily growing. Applications include automotive (driver assistance and autonomous vehicles), robotics, security, and medical imaging systems.
Two questions arise when designing camera systems for such applications.
- How best to select (or qualify) cameras for MV/AI applications?
- What image processing (ISP or filtering) is optimal?
To answer these questions, we must go beyond standard measurements of sharpness (MTF) and noise and apply metrics derived from information theory, including information capacity and related metrics for object and edge detection.
These metrics are important because Object Recognition (OR), MV, and AI algorithms operate on information, not pixels, making them far better predictors of system performance than MTF or noise.
Imatest has developed highly convenient methods for measuring information capacity and related metrics. The white papers (with varying degrees of detail) describe how the new metrics can be used to select (or qualify) cameras and determine the optimum Image Signal Processing (ISP) for Object Recognition, which is likely to improve the performance of MV and AI algorithms.
Video: Image Information Metrics in Imatest
White papers
Three November 2023 white papers (probably too many; it may be reduced to two) contain the latest information capacity measurements and results,
|
The three white papers linked below contain the same essential material.
|
Camera Information Capacity from Siemens Stars (2020) describes a method that is slower and less versatile than the slanted edge, but better for observing the effects of image processing artifacts such as demosaicing, data compression, etc.
Try the Latest Metrics in the Imatest Pilot Program
Imatest 24.1 Alpha includes Edge SNRi, filter design, and many other significant enhancements over the 23.2 release.
Join the Imatest Pilot Program to try out these metrics in Imatest 24.1 Alpha.
Imatest 24.1 Beta will be available in early Spring. The full release of Imatest 24.1 will be in late Spring 2024.
Imatest Software Documentation
- Information capacity measurements from Slanted edges: Equations and Algorithms (2023) – figures of merit that combine sharpness and noise, conveniently measured from any slanted-edge, including NPS, NEQ, and SNRi. Similar to the white papers, which will be kept more up-to-date.
- Information capacity measurements from Slanted edges: Instructions (2023) – instructions on the new calculations. Incomplete as of late January, 2024.
- Shannon information capacity from Siemens stars This calculation from 2020 needs to be reviewed and updated. Some corrections may be needed.
Meaning – Results – Summary
Improvement to the ISO 12233 slanted-edge MTF (SFR) equation
In the process of calculating the new information metrics, we found and fixed a problem with ISO 12233 MTF algorithm that caused artifacts to appear at high frequencies. This involved interpolating scan lines prior to binning.
Improved slanted-edge MTF calculation – We found errors in calculations that depended on MTF2(f)/NPS(f). They were reduced by an improvement in the ISO 12233-based MTF calculation.
Electronic Imaging 2024
Norman Koren presented a keynote talk (one hour time slot) on Image Information Metrics (covering much of the material in the white papers) at Electronic Imaging 2024, January 21-25, 2024 at the Hyatt Regency hotel in Burlingame, California (near SFO). The talk was held at\ 8:45AM, Monday, January 22.
The slides from the talk (PDF format) are available here.
I will be writing a technical paper with the contents of the talk that will be published on imaging.org, due February 15. It will be linked from this page.
International Standard (in development)
ISO 23654 began development in early 2023. It is based on ISO 12233 and defines how to calculate information capacity from a slanted edge. It is in the PWD (preliminary working draft) phase.
Call for Participation
We are interested in working with machine vision and computer vision experts who are experienced with studying how image quality metrics correlate with object detection performance in computer vision systems. Please contact infocap@imatest.com if you are interested.