Research on Methods of Image Recognition for LCD Digit
Abstract
The old material weighing system using LCD liquid crystal display weighing data, the system intact, but no data export interface. Without affecting the normal use of equipment, the machine vision technology was used to achieve weighing data identification. In this paper, the digital image of LCD liquid crystal display was taken as the object of study, and the methods of seven-segment LCD digital image recognition were studied. 100 LCD display images were collected using the visual system. Two kinds of recognition methods were contrastively studied. One was area feature recognition method.0-9 character images were obtained through region feature method, and the numbers were identified according to these features. The other was the template matching method. Under the same condition, the results showed that the accuracy of data recognition is 100% for 100 images processing and 94% for template matching. Changing the condition, template matching method could also get the same accuracy. This shows that the regional feature method and template matching method can meet the actual identification requirements.
Keywords
template matching; area feature; machine vision; digit recognition; binarization
DOI
10.12783/dtetr/iceeac2017/10747
10.12783/dtetr/iceeac2017/10747
Refbacks
- There are currently no refbacks.