<p><strong>Machine vision techniques for identifying the symptoms of <em>Tetranychus urticae</em> (Acari: Tetranychidae), <em>Helicoverpa armigera</em> (Lepidoptera: Noctuidae), and calcium deficiencies in tomato plant</strong></p>

Accurate detection and identification of damaging factors are key requirements for their effective management and control. Conventional methods for detecting these pests in plants require continuous monitoring, which is a costly and time-consuming process on farms and is accompanied by unavoidable human errors. Therefore, it is necessary to use modern technologies to help farmers automatically and promptly identify key pests. In this study, a convolutional neural network with Inception_v3 architecture was used to automatically detect damage symptoms in tomato plants. A Sony DSC-WX200 camera with an effective sensor resolution of 18 megapixels was used to collect images of the symptoms caused by the pests. To evaluate the performance of the convolutional neural network with Inception_v3 architecture, the parameters of average precision, precision, and recall were used. The results showed the accuracy of the Inceptionv3 architecture in identifying the symptoms of T. urticae (99.47%), H. armigera (99.18%), and calcium deficiency (99.77%). The results indicate that our proposed system, which uses recorded images, can provide an efficient and accurate solution for identifying damaging factors in tomato plants.