*Result*: Application of Image Big Data in the Ecological Agricultural Product Supply Chain.

*With the rapid development of ecological agriculture and the increasing demand for agricultural product supply chain management, effectively monitoring the quality and circulation status of agricultural products has become an urgent issue. Image big data technologies, particularly advancements in deep learning and computer vision, offer innovative solutions for surface quality detection, analysis, and traceability of agricultural products. By precisely estimating surface disparity and analyzing quality, these technologies not only improve the efficiency of quality control but also enhance supply chain transparency, ensuring the stability of product quality. However, existing image analysis methods face significant limitations when dealing with minor surface defects, lighting variations, and complex textures. Traditional image processing techniques are less effective in these areas, and the application of deep learning is still in the exploratory phase. To address these issues, this study proposes a deep learning-based method for surface disparity estimation of agricultural products and designs three innovative models: 1) a Convolutional Neural Network (CNN) for surface disparity estimation of agricultural products, 2) an end-to-end deep learning stereo matching model for surface disparity estimation, and 3) a deep learning pyramid stereo matching network model for surface disparity estimation of agricultural products. These models aim to overcome the shortcomings of current methods and enhance the precision and stability of agricultural product image analysis, providing more efficient and intelligent technical means for quality control in the agricultural product supply chain. [ABSTRACT FROM AUTHOR]

Copyright of Traitement du Signal is the property of International Information & Engineering Technology Association (IIETA) and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)*