Noisy Image Meter Reading via Dynamic Programming
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current automated meter reading technologies face challenges in accurately interpreting meter readings from noisy images, leading to incorrect data and high costs associated with upgrading to smart meters for real-time monitoring and billing.
Innovation Solution
A system and method that utilize likelihood arrays and chain sets to assemble and splice paths from sequences of meter images, incorporating chain criteria and selection criteria to produce a stable and statistically consistent final trajectory of meter readings, enabling reliable data collection without the need for expensive infrastructure upgrades.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If automated meter reading is implemented using legacy meters with image capture, then cost is reduced compared to smart meter replacement, but measurement precision deteriorates due to noisy image data
Solution Approach 1:
The system performs preliminary actions by capturing multiple images of the meter readout at different times before final interpretation. These preliminary captures create a sequence of observations that are processed through dynamic programming to establish the most likely true reading, accounting for potential noise or errors in individual images.
Solution Approach 2:
The system uses feedback by comparing the captured image data against expected meter reading patterns and using dynamic programming to evaluate multiple possible interpretations. The system iteratively refines the reading by considering the sequence of observations and selecting the trajectory with the highest probability, effectively using feedback from the data itself to improve accuracy.
2Device complexity
If simple image recognition is used for meter reading, then device complexity is reduced, but reliability deteriorates due to incorrect readings from noisy images
Solution Approach 1:
The system segments the meter reading problem into discrete time steps and digit positions, creating a grid of possible readings. Each cell in this grid represents a possible digit value at a specific time, and the system evaluates probabilities for each segment independently before combining them through dynamic programming to form the complete reading trajectory.
3Measurement precision
If multiple images are captured and processed through complex algorithms, then measurement precision improves, but loss of time increases due to processing requirements
Solution Approach 1:
The system applies dynamic programming, a dynamic optimization method that efficiently processes the sequence of meter images. By breaking down the problem into smaller subproblems and using memoization to avoid redundant calculations, the system can handle multiple images and complex probability calculations without excessive processing time, adapting the computation to the actual data quality and noise levels.
Data Source
AI summary
A meter readout on a meter has digits including a first digit, a second digit, etc. A sequence of images of the meter is obtained. The images include images of the digits in the meter readout. Automated recognition of the digits in the images result in likelihood arrays indicating the likelihoods for the digit values for the digits imaged in the meter images. Short chains of digit values are identified and spliced together to form a series of single digit, two-digit, three-digits, etc. paths that are built up based on the likelihood arrays. Various criteria are used to discard most of the chains and thereby avoid the combinatorial explosion of possible paths and thereby produce reliable meter readings without consuming considerable computational resources.