Implementing computer vision technology and innovative practices in photography education
With computer vision (CV) technology, students may get immediate comments on their photos' exposure, lighting, and composition, which boosts both their technical skills and their ability to experiment creatively. To evaluate lighting, focus, composition, and other aspects of photography with the use of deep learning (DL) models, this study introduces a novel framework that integrates automated picture analysis with tailored suggestions. An Intelligent Jellyfish Optimized Residual Neural Network (IJFO-ResNet) handles image recognition and classification, with K-means clustering employed for technical attribute-based categorization. For training and evaluation, the model uses a broad dataset that covers several genres of photography. To ensure the data is intact, it incorporates preprocessing techniques such as noise reduction and cleaning low-quality photographs. Students can improve their photography abilities through the use of DL algorithms that generate feedback, which in turn provides them with practical insights and opportunities for interactive learning. Metrics for evaluation include the precision of technical comments, student participation, and the degree to which abilities are improved as shown by pre- and post-tests. In picture recognition and classification tests, the suggested model had relatively high performance with 98% accuracy, 92% precision, 94% recall, and 96% F1-score. The results show that the model is able to accurately evaluate the three main characteristics of any photograph: exposure, clarity, and composition. When used in conjunction with educational or training settings, it can automatically provide students with constructive criticism on their uploaded photos, allowing them to hone their photography skills through practice and the application of any necessary adjustments. These collective findings point to a fresh approach to divergent personal learning that is procedurally orientated; this approach will provide both students and teachers with intelligent resources to expand upon, fostering the development of abilities and creativity.
Authors
- Hongbin Zhou (ORCID: https://orcid.org/0000-0002-2824-2277)
Institutions
- Jimei University (CN)
Publication Details
- Journal
- Systems and Soft Computing
- Published
- 2025-11-21
- DOI
- https://doi.org/10.1016/j.sasc.2025.200419
- Primary Topic
- Participatory Visual Research Methods
- Type
- article
- Field-Weighted Citation Impact
- 0.00