Evaluation of multispectral imaging for freeze damage assessment in strawberries using AI-based computer vision technology

Authors: SUNIL, GC - North Dakota State University; KHAN, AMIN - North Dakota State University; Horvath, David; SUN, XIN - North Dakota State University

Submitted to: Smart Agricultural Technology
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 2/19/2025
Publication Date: 2/21/2025
Citation: Sunil, G., Khan, A., Horvath, D.P., Sun, X. 2025. Evaluation of multispectral imaging for freeze damage assessment in strawberries using AI-based computer vision technology. Smart Agricultural Technology. https://doi.org/10.1016/j.atech.2025.100851.
DOI: https://doi.org/10.1016/j.atech.2025.100851

Interpretive Summary: Quantifying the amount of damage caused by freezing in strawberries is difficult, slow, and subjective. To overcome this problem, a series of deep learning models were trained and tested on a diverse set of images of strawberry plants that had variable damage levels following different preliminary growth conditions and freezing treatments. The models were tasked with identifying the differences between the plant leaves and the background, and then were trained to classify individual pixels on the leaves as either dead, damaged, or undamaged. One model (MobileNetV3) proved to be the best for this task. The results of this work will allow the development of a simple application to quantify freezing damage in strawberries using high-throughput methods with limited biases.

Technical Abstract: Traditional freezing damage quantification on plants often relies on human visual assessment, which can be inconsistent between individuals and even for the same person at different times. Additionally, this approach is time-consuming and labor-intensive. A computer system and application utilizing computer vision-based freezing-damage quantification artificial intelligence models have the potential to significantly streamline and improve the consistency of freezing damage assessment. This study focuses on the development of lightweight deep learning models with reduced numbers of parameters for freezing damage quantification. These models utilize semantic segmentation techniques to accurately classify and quantify background, dead tissue, and healthy tissue in plant images. Several deep learning image segmentation architectures, including DeepLabv3, UPerNet, PSPNet, U-Net, and SegNet, along with modification of UperNetResNet101 with attention mechanism and backbone architectures (densenet, efficientnet, convNeXt, and mobilenetv3) were trained on a greenhouse collected strawberry dataset. The UperNet model modified with a DenseNet backbone achieved the highest mean Intersection Over Union value of 0.77, surpassing the performance of the base model (0.76). Notably, a lightweight model (UperNetMobileNetV3) with 93% fewer parameters than the UperNetResNet101 model demonstrated a performance degradation of only 1.76%. This performance improvement can be attributed to a combination of factors, including variations in model complexity and the limited number of training samples. DenseNet architectures, known for their ability to efficiently reuse features, demonstrated strong performance. Additionally, MobileNetV3, an architecture specifically designed for edge devices, proved to be surprisingly effective for the complex task of freezing damage segmentation. Overall, testing of this AI-based model demonstrated its potential for freezing damage quantification in plant species beyond strawberries.