Page 357 - AI Ver 3.0 class 10_Flipbook
P. 357

• By reducing the spatial dimensions, pooling helps lower the risk of overfitting by simplifying the model,
                   ensuring it generalises well to new, unseen data.






                                                                                       Max

                                                                          Pooling





                                                                                      Sum





                 A small difference in the input image will create a very similar pooled image.

                 Fully Connected Layer


                 This is the last and the final layer of the Convolutional Neural Network. After the features of the input image
                 are extracted by the convolution layers and downsampled by the pooling layers, their output is a 3-dimensional
                 matrix which is flattened into a vector of values. These values of the single vector represent a specific feature of a
                 specific label and are redirected to fully connected layers to predict the final outputs of the network. This helps in
                 classifying an image into a specific label based on the probability of the input being in a specific class. For example,
                 the input image with a beak belongs to the category of a bird.

                                                                                                        bird    p bird

                                                                                                       sunset   p sunset


                                                                                                        dog      p dog

                                                                                      ...                        p
                                                                                                        cat       cat
                      Convolution + Nonlinearity     Max pooling             Vec                       ...


                                 Convolution + Pooling layers                Fully connected layers Nx binary classification
                                                      Image Classification Using CNN




                              Brainy Fact


                        The first image recognition CNN-AlexNet named after(creator-Alex Krizhevsky), won the 2012 ImageNet
                        Computer Vision contest with 85% accuracy.









                                                                                    Computer Vision (Practical)  355
   352   353   354   355   356   357   358   359   360   361   362