Artificial Neural Networks for Image UnderstandingA tutorial of theoretical and practical principles of neural networks as applied to complex computing tasks such as robotics vision control, medical image analysis and remote sensing. The rapidly-evolving concept of automated image understanding is fully explored and explained in real-world terms, with numerous examples of industrial and commercial applications. |
Contents
Preprocessing | 13 |
Feature Extraction | 49 |
Texture Analysis | 75 |
Copyright | |
7 other sections not shown
Common terms and phrases
algorithm analog ANN models artificial neural network back-propagation bidirectional associative memories character recognition Chellappa classifier cluster coefficients competitive learning Computer Conference on Neural connection corresponding data compression Digital Image Digital Image Processing edge feature extraction feature space feature vector feed-forward Fourier transform Gabor filters given gray level gray value Grossberg Hopfield network IEEE IEEE Transactions Image Processing image restoration implementation input image input layer input pattern input vector International Joint Conference interpolation invariant iterations Kohonen Kulkarni and Byars L₂ layer L₁ linear matching methods motion parameters neural network architecture neural network models neurochip Neurocomputing neuron noise observation vector obtained optical flow original image output layer pattern recognition perceptron permission pixel Proceedings of International processor Radon transform represents rotation samples San Diego segmentation shown in Figure spatial stereo synapses techniques tion units in layer unsupervised updated vector vision w₁ weights between layers Zhou Σ Σ