발간년도 : [2024]
| 논문정보 |
|
| 논문명(한글) |
[Vol.19, No.3] A Study on the Use of Image Processing Classes in Generative AI |
|
| 논문투고자 |
Jee-Yeon Lee, Jae-Woong Kim |
|
| 논문내용 |
Generative AI is being utilized in various practical fields to enhance work efficiency. Particularly in programming education, it is employed to overcome individual differences and amplify the educational effects through code generation. This study focuses on exploring the possibility and effective use of code generation using ChatGPT in projects related to image processing using OpenCV. The research methodology includes using ChatGPT to generate code during project classes and applying it in practice. The process involves obtaining prompts for restoring old photos and required algorithms, then receiving detailed implementation code in response, and applying it to the photos. Gaussian and bilateral blurs, as well as sharpening techniques, are employed. In cases where applying the obtained code worsened the damage, the process involved iterating with additional prompt inputs to address the issues. When detailed damage occurred even after filtering, suggestions for additional operations, such as histogram equalization in OpenCV for contrast adjustment, were proposed to achieve the desired results. The research findings indicate that ChatGPT effectively generates code based on prompts, allowing learners to experience problem-solving and code optimization. However, there are challenges with generative AI, such as the need for repetitive prompt inputs and adjustments to obtain desired results. This study demonstrates the positive utilization of generative AI, empowering learners to generate effective code and resolve issues through iterative prompt inputs and adjustments. |
|
| 첨부논문 |
|
|
|
|
|
자료실 > 논문지 
