Images can be deteriorated for a variety of reasons. For instance, blurry images are produced by out-of-focus optics, while noise is produced by variations in electrical imaging systems. The blurred image classification and deblurring approach provided here uses DWT. After classifying the blurry image, there are ways to deblurring the given blurry image. The aim of blur image classification is finding blurred or unblurred images from input images. The deblurring of the photographs is presented toward the conclusion. This suggested deblurred image categorization and deblurred image can produce the best results. Finally, we evaluate the parameter analysis. This proposed scheme uses textural feature-based image classification using a neural network using a machine learning approach. Texture features are removed using the Gray level co-occurrence matrix, and the artificial neural network is advanced for the classification of images into dissimilar classes.