Retail Weighing System Image-Based Container Weight Compensation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In retail settings, weighing systems struggle to accurately measure the weight of commodities when both weighed and unweighed items are placed on the scale simultaneously, leading to incorrect weight calculations and pricing.
Innovation Solution
A weighing system comprising a weighing section, an image capturing section, an identification module, and calculation modules that measure and calculate the weight of unmeasured commodities by identifying container weights from captured images and commodity information, allowing for precise pricing calculations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If both weighed and unweighed commodities are placed on the weighing device simultaneously, then the weighing device measures the total weight of all commodities, but it cannot distinguish between weighed and unweighed commodities, leading to incorrect weight measurement of the unweighed commodity
Solution Approach 1:
The system segments the weight measurement process into three distinct components: container weight (identified from images), pre-weighed commodity weight (read from labels), and unweighed commodity weight (calculated by subtraction). This segmentation allows accurate measurement of unweighed commodities even when mixed with weighed commodities on the scale.
Solution Approach 2:
The system introduces an intermediary calculation process that uses image recognition to identify container weights and OCR to read pre-weighed commodity weights from labels. These intermediary measurements serve as reference values that enable accurate determination of unweighed commodity weights through mathematical subtraction from the total weight.
2Productivity
If the weighing device only measures total weight without additional information, then the device operation is simple, but it cannot calculate correct prices when both weighed and unweighed commodities are present
Solution Approach 1:
The system performs preliminary actions by capturing images of containers and their labels before the weighing process. Image recognition identifies container weights and OCR reads pre-weighed commodity weights from labels in advance. This preliminary information gathering enables efficient pricing calculations without adding complexity to the actual weighing operation.
Solution Approach 2:
The system implements feedback by using the total weight measurement from the scale combined with previously captured image and label data to calculate the weight of unweighed commodities. This feedback loop ensures accurate pricing information is generated even when customers accidentally place both weighed and unweighed commodities on the scale simultaneously.
Data Source
Figure 1
Figure 2~3
Figure 4
AI summary
A weighing system comprises a weighing section, an image capturing section, an identification module, a reading module and a first calculation module. The weighing section measures a weight of a container and a commodity in the container. The image capturing section photographs the container weighed by the weighing section and commodity information including information indicating the weight of the commodity in the container. The identification module identifies the weight of the container contained in the image captured by the image capturing section. The reading module reads the weight from the commodity information contained in the image captured by the image capturing section. The first calculation module calculates the weight of an unmeasured commodity from the weight measured by the weighing section, the weight in a category of the container identified by the identification module and the weight read from the commodity information by the reading module.