In today’s world, due to irregular climatic patterns and other environmental issues various pests will affect the crops. These issues may affect the soil nutrition too. Due to this deficiency in nutrition, several diseases may affect the crops. In large agricultural field farmers feel difficult to monitor the pest in every nook and corner. Before identifying the pests, it may spread over vast area and cause severe damage to crops. Also they may not aware of bacterial disease that may affect the crops and their symptoms. Besides farmers may not know which type of pesticide and chemicals they need to use to prevent the further damage of crops. In some cases excess usage of those pesticides and other chemicals may affect the corn fields. Improper usage of those chemicals may affect the yield. To monitor the field periodically we need more human power. These problems are overcome by our proposed system. In our system diseases/pests are identified at early stages by capturing the images periodically in the agricultural field. Using image processing techniques the captured image will be segmented. For this purpose Texture based Segmentation and Simple Linear Iterative Clustering (SLIC) are used. From the segmented images the features for identifying pests and diseases will be extracted. The extracted features are used for classification. The presence of pest/disease will be identified at the first stage. In second stage the type of pest/disease will be detected. For classification both Binary Support Vector Machine (BSVM) and Multi class Support Vector Machine (MSVM) are used. And also, the pesticides and other chemicals which are needed to protect the field from further damage will be recommended along with the amount of chemicals needed and the method of usage of those chemicals.

在当今世界，由于不规则的气候模式和其他环境问题，各种害虫都会影响农作物。这些问题也可能影响土壤营养。由于营养不足，几种疾病可能会影响作物。在大型农业领域中，农民很难在每个角落和角落监控害虫。在识别有害生物之前，它可能散布在广阔的区域，并严重损害农作物。他们也可能没有意识到可能影响农作物及其症状的细菌病。此外，农民可能不知道他们需要使用哪种农药和化学物质来防止农作物的进一步损害。在某些情况下，过量使用这些农药和其他化学物质可能会影响玉米田。这些化学药品使用不当可能会影响产量。要定期监视现场，我们需要更多的人力。我们提出的系统克服了这些问题。在我们的系统中，通过在农业领域中定期捕获图像，可以在早期识别出疾病/虫害。使用图像处理技术，将对捕获的图像进行分割。为此，使用了基于纹理的分割和简单线性迭代聚类（SLIC）。从分割的图像中，将提取出用于识别病虫害的特征。提取的特征用于分类。在第一阶段将确定有害生物/疾病的存在。在第二阶段，将检测有害生物/疾病的类型。为了分类，使用了二进制支持向量机（BSVM）和多类支持向量机（MSVM）。并且，