Low-Resolution Tactile Image Recognition for Automated Robotic Assembly Using Kernel PCA-Based Feature Fusion and Multiple Kernel Learning-Based Support Vector Machine

Figure 10

An illustrative example for the image preprocessing stage, where (a) is the testing object, (b) is the corresponding 10-by-10 tactile image, (c) is the resized 33-by-33 image, (d) is the image after Gamma correction-based contrast enhancement, (e) is the image after noise reduction, and (f)–(h) are the binarized images with 1×1, 2×2, and 3×3 image partitioning-based thresholding.

(a)  

(b)

(c)

(d)

(e)

(f)

(g)

(h)