Quantitative Image Quality Assessment for Photogrammetry Texture
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing photogrammetry tools lack the ability to quantify the quality of surface texture in objects before image capture, leading to poor photogrammetry results due to inadequate texture, which is often discovered post-processing and not addressed until after image acquisition.
Innovation Solution
A system and method for quantitative image quality assessment that evaluates the spatial and frequency content of images to compute image quality metrics, providing a Pass/Fail determination, image quality maps, and recommendations for improving texture, deployable on various devices and platforms, to assess and enhance image texture before photogrammetry processing.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If photogrammetry processing is performed without pre-assessment of surface texture quality, then the photogrammetry workflow can proceed quickly, but the results will be poor quality with missing regions and rough surface modeling
Solution Approach 1:
The patent applies preliminary action by performing image quality assessment before photogrammetry processing. The system analyzes images for surface texture quality, lighting conditions, and feature detectability prior to the main photogrammetry workflow, allowing users to identify and correct issues like poor texture or inadequate lighting before they result in failed reconstructions. This pre-assessment prevents time loss from discovering quality issues after processing.
2Reliability
If existing photogrammetry tools are used without image quality assessment, then the tool complexity remains low, but the reliability of photogrammetry results deteriorates due to undetected poor surface texture
Solution Approach 1:
The patent applies segmentation by dividing the photogrammetry workflow into distinct stages: image quality assessment (analysis of texture, lighting, features) and photogrammetry processing. The assessment module independently evaluates image quality metrics and provides feedback, while the photogrammetry module processes images when quality thresholds are met. This segmentation increases reliability by ensuring quality checks are performed without significantly increasing overall system complexity.
3Productivity
If images with poor surface texture are processed, then image capture can be completed quickly, but the photogrammetry output will have rough surfaces and missing regions
Solution Approach 1:
The patent applies feedback by implementing an image quality assessment system that analyzes captured images for surface texture quality, lighting adequacy, and feature detectability. The system provides quantitative feedback metrics and visual overlays indicating poor quality regions, enabling users to retake images or adjust lighting/texture conditions before photogrammetry processing. This feedback loop maintains productivity by quickly identifying bad images while ensuring high surface modeling quality through corrective actions.
Data Source
AI summary
A system and method for quantitative image quality assessment for photogrammetry are disclosed. An example embodiment is configured to receive one or more quality assessment images via an image receiver; process the quality assessment images to determine if a quality of the surface texture features of an object or environment depicted in the quality assessment images satisfies a pre-determined quality threshold, the pre-determined quality threshold corresponding to a likelihood of a satisfactory result if images of the object or environment are used for photogrammetry; generate an image quality map indicating image quality values corresponding to regions of the surface of the object or environment that have satisfactory or unsatisfactory texture features for photogrammetry; and generate instructions or prompts for a user, the instructions or prompts directing the user to perform actions with respect to the object or environment that will effect improvements of the texture features for satisfactory photogrammetry.