Massive image data was generated under the background of big data. Intelligent analysis, processing, and identification of the image information of the agricultural IoT terminal can intuitively and vividly express all aspects of crop growth, development, health status, damage degree, etiology, etc. The system can respond similarly to intelligent living beings. The current image recognition technology for crops is relatively backward in terms of image recognition accuracy and time. In this case, among the millions of data at every turn, the data processing capacity is far from enough, so this paper proposes based on Smart agriculture uses deep learning video image analysis technology to monitor system research, and through the research and improvement of the deep learning video image analysis technology monitoring system, it can be better applied to smart agriculture. This model is applied to the corn disease pictures collected from farmland to accurately identify corn diseases. The experimental results show that the overall recognition accuracy of major corn diseases (maize maize leaf spot, small leaf spot, gray spot disease, smut, and black powdery mildew) using the improved convolution NNs optimization algorithm reaches 93.2% Compared with a single convolutional NNs, the accuracy is improved by a quarter, and the processing time of each picture is shortened by about one-tenth that of the traditional NNs. The accuracy and training speed of this algorithm are significantly improved compared with traditional convolutional NNss.

Bixuan Wan and Jing Si. Monitoring System of Deep Learning Video Image Analysis Technology Based on Smart Agriculture. Academic Journal of Agricultural Sciences (2023), Vol. 4, Issue 1: 68-80. https://doi.org/10.38007/AJAS.2023.040106.

[1] Baker M A , Robinson J S.(2016). “The Effects of Kolb's Experiential Learning Model on Successful Intelligence in Secondary Agriculture Students.”, Journal of Agricultural Education, 3(2),pp.57-63.

[2] Cao L , Meng Y , Lu J , et al.(2015). “Research on Video Image Recognition Technology of Maize Disease Based on the Fusion of Genetic Algorithm and Simulink Platform.”, Springer International Publishing,12(1),pp.147-155.

[3] Ramcharan A , Baranowski K , Mccloskey P , et al.(2017). “Using Transfer Learning for Image-Based Cassava Disease Detection.”, Frontiers in Plant Science,13(9),pp.89-96.

[4] Kumar S , Kaur R .(2015) “Plant Disease Detection using Image Processing- A Review.”, IJCA,124(2),pp.528-535.

[5] S.W. Zhang, Y.J. Shang, L. Wang.(2015). “Plant disease recognition based on plant leaf image.”, Journal of Animal & Plant Sciences, 25(3),pp.42-45.

[6] Ma J , Du K , Zheng F , et al.(2018). “Disease Recognition System for Greenhouse Cucumbers Based on Deep Convolutional NNs.”, Nongye Gongcheng Xuebao/Transactions of the Chinese Society of Agricultural Engineering, 34(12),pp.186-192.

[7] Hong, Seunghoon, You, Tackgeun, Kwak, Suha.(2015). “Online Tracking by Learning Discriminative Saliency Map with Convolutional NNs.”,Computer Science, 43(13),pp.597-606.

[8] Zeng Haoyang, Edwards Matthew D, Liu Ge.(2016). “Convolutional NNs Architectures for Predicting DNA–protein Binding.”, Bioinformatics, 25(11),pp.12-20.

[9] Vinayak R , Lizhen L , Dunson D B.(2016). “Data Augmentation for Models Based on Rejection Sampling.”, Biometrika, 13(2),pp.2-10.

[10] Rosario R R.(2017). “A Data Augmentation Approach to Short Text Classification.”, Proquest Llc, 17(4),pp.267-275.

[11] Jayaram, Vinay, Alamgir, Morteza, Altun, Yasemin.(2015). “Transfer Learning in Brain-Computer Interfaces.”, IEEE Computational Intelligence Magazine, 11(1),pp.20-31.

[12] Rojas-Carulla, Mateo, Schölkopf, Bernhard, Turner, Richard.(2015) “Invariant Models for Causal Transfer Learning.”, Statistics, 15(2),pp.67-73.