876ce96a58 Add datasets/fake/generate_dataset.py
Inability to reproduce IPOL 77777000278 demo
408b7a2ba9 Add flat vision dataset size computation
Benjamin_Loison commented on issue Benjamin_Loison/Robust_image_source_identifica…#9 2024-03-21 16:25:07 +01:00
Context-Adaptive Interpolator (CAI)

Now that have possibly a first version, should try to compare with other models to see if it works.

e23543506a Correct getPixelIndexesAround missing one pixel
13de984d59 Add articles/Wiener_Filtering/
Benjamin_Loison commented on issue Benjamin_Loison/Robust_image_source_identifica…#12 2024-03-21 15:38:00 +01:00
How to manage the border of the image when need neighborhood?

Pad with black make same image size, easing comparing them.

Benjamin_Loison commented on issue Benjamin_Loison/Robust_image_source_identifica…#10 2024-03-21 15:34:31 +01:00
Wiener filter
from PIL import Image, ImageChops
import math
import operator
import functools
import matplotlib.pyplot as plt
import numpy

def openImage(filePath):
    return Image.open(filePath).
Benjamin_Loison commented on issue Benjamin_Loison/Robust_image_source_identifica…#10 2024-03-21 15:15:56 +01:00
Wiener filter

With Brightness 127 and Contrast 100, get:

Before Wiener filter:

image

After Wiener filter:

![image](/attachments/1ad5f8fd-…

Wiener filter
f3e96243f3 Fix #10: Implement Wiener filter
739317fa63 WIP: Context-Adaptive Interpolator (CAI)
d43ff5777f Add articles/Wu_ICIP_2012/
Benjamin_Loison opened issue Benjamin_Loison/Pillow#3 2024-03-21 11:52:01 +01:00
ImagePixels[0, :] would ease naming neighbor pixels
Benjamin_Loison opened issue Benjamin_Loison/Pillow#2 2024-03-21 11:24:32 +01:00
PixelAccess - PixelAccess to make an image difference
How to manage the border of the image when need neighborhood?
45726968f2 Fix #12: Use a new image (#15)
Benjamin_Loison opened issue Benjamin_Loison/Pillow#1 2024-03-21 10:50:12 +01:00
Add - support to Image.crop
Benjamin_Loison created repository Benjamin_Loison/Pillow 2024-03-21 10:49:59 +01:00
Accelerate computation