| Summary: | 碩士 === 國立臺灣科技大學 === 自動化及控制研究所 === 98 === In recent years, Light Emitting Diodes (LED) display panels are commonly used in outdoor billboards, display screens, traffic indicators, scrolling signs, etc. LED display panels offer many advantages which include high color rendering, self-emission, wide viewing angle, fast response time, high luminous efficiency, low driving-voltage, long life, mercury-free (eco-friendly) attribute etc. The future of LED display panels looks bright. However, the production and inspection technology for a color LED module is still under development. To improve the manufacturing efficiency for LED modules, therefore, an automated inspection system for color quality is required. The main purpose of this study is to develop an automated color inspection and compensation solutions for color LED modules which are main components of a color LED display panel. In this study, an automated optical inspection technology and a LED PWM driving method were combined to inspect and improve the color quality for each pixel within a LED module. Initially, a self-developed color optical sensing module and a colorimeter were used to measure the tri-stimulus of LED pixels. After the color matching functions from the collected data was corrected by least-squares approximation method, the calibrated measured sensing module was applied to inspect the color characteristics of LED modules. And the chrominance and luminance data are well analyzed to perform a white light chrominance correction and luminance uniformity compensation for LED display modules. A Compensated Look Up Table (CLUT), built in a hardware driving circuit, as a result, was used to store the LED module’s compensation parameters. Some 2R1G1B LED modules (16×16 LEDs) from a local industry were verified in this study. The experimental results from the central 8×8 LEDs show that the average color difference of a target D65 white light with a chrominance of (0.1978, 0.4683) by using a 2R1G1B LED module is about 0.027006 beforce compensation, but the average color difference can be reduced to 0.003474 after it was compensated by using the proposed method. Meanwhile, the lumi nance difference was also reduced to less than 3.541%. Therefore, it is hopeful that this technology could be helpful for local LED industries to increase their manufacturing quality and to enhance their competition capacity as well.
|
|---|
