Flat Surface Detection in Photographs Using Autofocus Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for verifying the authenticity of photographs, such as those described in co-pending applications, are vulnerable to tampering and require multiple images or unreliable data connectivity, affecting user experience and efficiency.
Innovation Solution
A method using a digital camera's autofocus routine to record focal length values and input them into a machine learning classifier trained to distinguish between 'flat' and '3D' scenes, providing a fast and non-disruptive indication of a photograph's authenticity, which can be implemented on consumer devices like smartphones and potentially verified on a remote server.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If multiple photographs are taken at different exposures to verify authenticity, then the reliability of verification is improved, but the ease of operation deteriorates due to requiring two photographs instead of one
Solution Approach 1:
The patent extracts the verification process from requiring multiple photographs to using a single photograph combined with autofocus routine data. The autofocus routine, which already operates during normal camera operation, is repurposed to collect focal length information that serves as a verification marker, eliminating the need for additional capture steps.
Solution Approach 2:
The autofocus routine is given a dual function: it continues to perform its primary role of achieving proper focus while simultaneously collecting focal length data for authentication purposes. This multi-functionality allows verification without adding separate operational steps.
2Measurement precision
If multiple photographs are sent to a server for processing, then the measurement precision of verification is improved, but the loss of time increases due to data transmission requirements
Solution Approach 1:
The patent extracts only the essential verification data (focal length information from autofocus routine) rather than transmitting entire photographs. This minimal data extraction maintains verification capability while dramatically reducing transmission time and data volume.
Solution Approach 2:
The autofocus routine collects focal length data during the normal photographing process itself, performing the verification data collection in advance. This preliminary action means verification data is ready immediately without requiring separate processing steps.
3Reliability
If flash is activated to take one of two photographs, then the reliability of scene capture is improved, but the use of energy increases
Solution Approach 1:
The patent extracts verification capability from the photographing process itself using existing autofocus mechanics, eliminating the need for additional energy-intensive flash activation. The autofocus routine provides sufficient verification data without requiring supplemental lighting.
Data Source
AI summary
A system and method is disclosed for detecting whether a photograph is of a flat surface, e.g. a “photograph of a photograph”, or of a real three dimensional scene. The method includes using a digital camera to take a photograph, and within a predetermined time period either before or after taking the photograph, recording focal length information from a plurality of focus areas using the autofocus routine of the digital camera. The recorded focal length information forms the input to a machine learning classifier which has been trained to classify scenes as “flat” or “3D”.