This is an implimented project with the help of reference from original papers Radiologist-Level Pneumonia Detection on Chest X-Rays with Deep Learning and Pneumonia Detection in chest X-ray images using Convolutional Neural Networks and Transfer Learning.you can also check my project report here
Please Collect the data set from kaggle-Chest X-Ray Images (Pneumonia).
The sample Image is shown below.
Images - > Preprocesing{data augumentation,Contrast ENhancement} ->Vgg16{Model}->classification results
It includes image rotation,flipping and addtion of gaussian noise.The data set size is increased to good scale,the results are shown below.
Gamma Contrast Enhancement: It is is a nonlinear operation used to encode and decode luminance or tristimulus values in video or still image systems. Mathematically explined by the following formula.
Gaussian Contrast Enhancement: Its a Non-linear transformation function to enhance brightness and contrast.
Brightness enhancement It is a linear trasnformation which adjust the brightness of the image based on the given gain value,after that the image also undergoes thresholding operation.
| Name | Value |
|---|---|
| Batch size-training | ` 128 |
| Batch size-validation | ` 128 |
| Batch size-testing | ` 8 |
| Number of epochs | ` 10 |
| Learning Rate | | 2e-4 |
The final classification results are shown in below table.For more information please check the my project report here
