Crop waterlogging image classification detection and implementation method based on Hadoop

A crop waterlogging image classification detection and implementation method based on Hadoop is characterized by comprising the following steps: a) collecting crop field waterlogging images, and performing image correction and preprocessing and image principal component analysis; b) uploading the image matrix to a Hadoop computing platform for distributed storage, and compiling a parallel neural network algorithm; and c) carrying out algorithm modeling and prediction, and carrying out classification analysis on crop waterlogging image information. Distributed parallel classification analysis of the image data under crop disaster stress is performed through the Hadoop framework so that modeling and prediction speed of the classification algorithm can be accelerated, and the larger the imagedata size is, the more obvious the advantage is compared with the single-machine mode. A neural network algorithm is compiled by using a Scala language, and the algorithm is suitable for parallel operation under a Hadoop fra