Tomato, as a globally important crop, its freshness directly affects postharvest quality, market value, and consumer acceptance. Traditional tomato freshness evaluation mainly relies on manual inspection and experience-based judgment, which is time-consuming, labor-intensive, and inefficient. Meanwhile, plasma technology has shown promising potential in agricultural preservation due to its safety and effectiveness, making the evaluation of tomato freshness after plasma treatment particularly important. In recent years, with the rapid development of deep learning technology, non-destructive detection methods based on image analysis have become important tools for agricultural product quality assessment. This study proposes an improved YOLOv8n-based model (named CFL-YOLOv8n) for tomato freshness detection. The method optimizes the C2f module by introducing a RetBlock residual structure, adopts an FDPN-DASI feature pyramid for cross-scale feature fusion, and incorporates the Focaler-IoU loss function to enhance bounding box localization accuracy. Additionally, a lightweight LSCD detection head is proposed to replace the original module to reduce parameters. It shows that the improved CFL-YOLOv8n model performs better performance for tomato freshness detection than comparison models, achieving mAP@.50 and mAP@.50:.95 of 90.0% and 88.5%, improved by 3.4% and 3.4% compared to the original YOLOv8n model, while significantly reducing parameters and computational complexity. The ablation experiments confirmed the effectiveness of each improved module. The proposed method provides an efficient solution for tomato freshness detection and offers technical support for evaluating the preservation effects of plasma treatment under controlled experimental conditions. Future work will focus on expanding the dataset scale and validating the model under more complex real-world environments.

Complex plasma widely exists in thin film deposition, material surface modification, and waste gas treatment in industrial plasma processes. During complex plasma discharge, the configuration, distribution, and size of particles, as well as the discharge glow, strongly depend on discharge parameters. However, traditional manual diagnosis methods for recognizing discharge parameters from discharge images are complicated to operate with low accuracy, time-consuming and high requirement of instruments. To solve these problems, by combining the two mechanisms of attention mechanism (strengthening the extraction of the channel feature) and shortcut connection (enabling the input information to be directly transmitted to deep networks and avoiding the disappearance or explosion of gradients), the network of squeeze and excitation convolution with shortcut (SECS) for complex plasma image recognition is proposed to effectively improve the model performance. The results show that the accuracy, precision, recall and F1-Score of our model are superior to other models in complex plasma image recognition, and the recognition accuracy reaches 97.38%. Moreover, the recognition accuracy for the Flowers and Chest X-ray publicly available data sets reaches 97.85% and 98.65%, respectively, and our model has robustness. This study shows that the proposed model provides a new method for the diagnosis of complex plasma images and also provides technical support for the application of plasma in industrial production.