Page 171 - Handbook of Deep Learning in Biomedical Engineering Techniques and Applications
P. 171

160   Chapter 6 Plant leaf disease classification based on feature selection







































                                    Figure 6.4 An example of rescaling and center alignment for a leaf image:
                                    (A) original image; (B) corresponding image with bounding box; (C) final
                                    rescaled and center-aligned image.

                                       Due to various contrasts in leaf region, the contrast enhance-
                                    ment method in Refs. [13] is used to change pixel intensities that
                                    benefit in case of providing more information in some areas of an
                                    image. The image data set will be divided into two subsets:
                                    training set and test set. Finally, the convolution neural network
                                    is applied to classify the given images.
                                       Many contrast enhancement methods have been widely
                                    applied to improve the quality of the image [12]. In this chapter,
                                    to ameliorate features that are low contrast to achieve improve-
                                    ment in terms of contrast quality, we use a contrast enhance-
                                    ment approach in Ref. [13] before further analysis. The main
                                    idea of this method is to preserve the mean brightness of an
                                    input image during contrast adjustment in local regions. Firstly,
                                    the input image in RGB color channels is converted into HSI
                                    ones. This approach only focuses on the intensity parameter
   166   167   168   169   170   171   172   173   174   175   176