BLNN: Multiscale Feature Fusion-Based Bilinear Fine-Grained Convolutional Neural Network for Image Classification of Wood Knot Defects

Figure 2

Original images of wood knot defect and those created through data augmentation: (a) original image, (b) vertical mirroring, (c) rotation by 180°, (d) horizontal mirroring, (e) adding Gaussian noise to image, (f) increasing the hue by 10, and (g) adding salt-and-pepper noise to the image.

(a)

(b)

(c)

(d)

(e)

(f)

(g)