Status : Verified
| Personal Name | Laurente, Gabriello M. |
|---|---|
| Resource Title | Automatic measurement of supermarket shelves using computer vision methods for efficient planning and monitoring |
| Date Issued | 11 January 2023 |
| Abstract | People have been continuously looking for ways to improve their daily lives using technology. One way to do this is by trying to automate activities that are repetitive and manually done. Supervisors of supermarket stores frequently perform a tiresome, repetitive, and time-consuming task of measuring different object sizes and space that they take up on a supermarket shelf. An automated approach that could assist supervisors in measuring an object's dimensions and area was proposed to address this problem. The proposal was presented as an alternative approach to manual and traditional measuring techniques. The time saved when measuring the dimensions of a product and the accompanying store shelf areas was the primary key performance indicator (KPI) for this research. As a secondary KPI, the accuracy of the object's dimensions and estimated store shelf area were evaluated. The proposed methodology made use of fiducial markers, specifically ArUco markers in trying to automatically measure shelf and product dimensions. The proposed solution was able to achieve a potential time saved of roughly around 2 hours and 46 minutes (41.24% time-save from the manual method) for a modern trade supermarket in Pampanga and approximately 2 days and 21 hours (58.33% time-save from the manual method) for modern trade hypermarket store in Metro Manila. Time saved (%) was measured by getting the time saved as a fraction of the reference time divided by the reference time and multiplied by 100% (reference time here is the manual time of measuring the shelves). Moreover, the potential time saved increases and scales with the size of the supermarket. The bigger the supermarket means more shelves and the greater time saved. |
| Degree Course | MS Industrial Engineering |
| Language | English |
| Keyword | automatic measurements; computer vision; fiducial markers; supermarket shelves |
| Material Type | Thesis/Dissertation |
Preliminary Pages
479.45 Kb
Category : F - Regular work, i.e., it has no patentable invention or creation, the author does not wish for personal publication, there is no confidential information.
Access Permission : Open Access