Research on bamboo strip density control technology based on deep learning.

Area of Science:

• Materials Science
• Computer Science
• Artificial Intelligence

Background:

• Traditional bamboo strip quality control relies on manual inspection, which is time-consuming and subjective.
• Bamboo density is a critical quality parameter, influenced by vascular bundle distribution.
• Automated methods are needed to improve the efficiency and accuracy of bamboo density assessment.

Purpose of the Study:

• To develop a deep learning-based method for automated bamboo strip density detection.
• To quantify bamboo density by analyzing vascular bundle distribution in cross-sectional images.
• To compare the performance of various deep learning models for this task.

Main Methods:

• A dataset of bamboo strip cross-sectional images was compiled.
• Eleven mainstream deep learning models, including Convolutional Neural Networks (CNNs) and Transformers, were trained and evaluated.
• Vascular bundle density classification was performed using these models.

Main Results:

• The ConvNeXt model demonstrated superior performance in vascular bundle density classification.
• The ConvNeXt model achieved a classification accuracy of 99%.
• Deep learning models effectively analyzed vascular bundle distribution for density quantification.

Conclusions:

• Deep learning offers an effective and automated solution for bamboo strip density control.
• The ConvNeXt model shows significant potential for precise bamboo quality assessment.
• This research highlights the applicability of AI in enhancing material quality inspection processes.