|
| 1 | +CNN |
| 2 | +========== |
| 3 | +* CNN are deep learning architectures that are primarily used for processing image data. |
| 4 | +* The special operation known as Convolution helps them extract features like edges and textures, in combination with filters. |
| 5 | +* ReLU is applied as a activation function to add non linearity |
| 6 | +* Pooling is perfomed to reduce the spatial dimensions while retainign important information. This is helpful in computational load and controlling overfitting |
| 7 | +* Fully Connected Layer (FCL) , after several convolution and pooling operations, the output is passes through a FCL to generate class probabilties needed for classification. |
| 8 | + |
| 9 | +How CNNs Work: |
| 10 | +========== |
| 11 | +* The input image is transformed into a numerical representation, where each pixel is assigned a value based on its intensity. |
| 12 | +* The convolution operation involves sliding the filter across the image and performing element-wise multiplication, followed by summation to create a feature map. |
| 13 | +* As data progresses through multiple layers, CNNs learn increasingly complex features, from simple edges in early layers to intricate shapes in deeper layers. |
| 14 | + |
| 15 | +Applications: |
| 16 | +========== |
| 17 | +CNNs are widely used in various fields such as: |
| 18 | +* Image Recognition: Identifying objects in images (e.g., facial recognition). |
| 19 | +* Medical Image Analysis: Analyzing X-rays or MRIs for diagnostic purposes. |
| 20 | +* Autonomous Vehicles: Object detection and scene understanding. |
0 commit comments