Exploring Blurriness in Image Stitching

My images have become blurred in certain areas after stitching. Could you please explain why this happens and which parameter can be adjusted to resolve it?

My code:

import os
import cv2
import numpy as np
from PIL import Image, ImageFilter
from stitching import Stitcher, AffineStitcher
from tqdm import tqdm 

panorama = None
for i in tqdm(range(0, len(image_files), dynamic_block_size), desc="processing"):
    block = image_files[i:i + dynamic_block_size]

    images = [cv2.imread(img_file) for img_file in block]

    if any(img is None for img in images):
        continue

    if panorama is None:
        if len(images) == 1:
            result = images[0]
        else:
            result = stitcher.stitch(images) 
    else:
        tmp_images = []
        tmp_images = [] + images + [panorama]
        result = stitcher.stitch(tmp_images)

    if result is None:
        continue
    else:
        panorama = result 

My settings parameters are as follows:

settings = {
"detector": "sift",
# "nfeatures": 1000,
"confidence_threshold": 0.5,
"crop": False,
"estimator": "affine",
"wave_correct_kind": "no",
"matcher_type": "affine",
"adjuster": "affine",
"warper_type": "affine",
"compensator": "no",
"medium_megapix": 0.8, # 0.5
"low_megapix": 0.5, # 0.1
"final_megapix": -1,
"range_width": -1,
"nr_feeds": 1,
"block_size": 32,
"try_use_gpu": True,
"blender_type": 'multiband',
"blend_strength": 1,
}

that bunch of config data doesn’t mean anything on its own.

what code are you running?

Hello, thank you for the reminder. I have updated my code.

because of that.

what happens when you copy a sheet of paper, and then take the copy and copy that again, and so on?

that is what happens.

Could you please provide me with some guidance on how to handle this situation? Each of my images is approximately 12MB in size, and when I attempt to concatenate multiple images directly into a list, I encounter memory errors with OpenCV. As a result, I have been employing a step-by-step concatenation approach. Thank you for your assistance!