ChatGPT로 쉽게 하는 이미지 처리 자동화

!pip install pillow

from PIL import Image, ImageEnhance
import os

def add_watermark(image_path, watermark_path, output_path, opacity=0.5):
    base_image = Image.open(image_path).convert('RGBA')
    watermark = Image.open(watermark_path).convert('RGBA')

    # Watermark 크기 조정
    width, height = base_image.size
    watermark_width, watermark_height = watermark.size
    scale_factor = 0.1  # 워터마크 크기를 원본 이미지의 10%로 조정
    new_watermark_width = int(width * scale_factor)
    new_watermark_height = int(watermark_height * (new_watermark_width / watermark_width))
    watermark = watermark.resize((new_watermark_width, new_watermark_height), Image.LANCZOS)

    # 투명도 조정
    alpha = watermark.split()[3]
    alpha = ImageEnhance.Brightness(alpha).enhance(opacity)
    watermark.putalpha(alpha)

    # 워터마크 위치 설정 (우측 하단)
    position = (width - new_watermark_width - 10, height - new_watermark_height - 10)

    # 워터마크 합성
    transparent = Image.new('RGBA', (width, height))
    transparent.paste(base_image, (0, 0))
    transparent.paste(watermark, position, mask=watermark)
    transparent = transparent.convert('RGB')  # 투명 배경 제거

    # 이미지 저장
    transparent.save(output_path)

def process_images(input_folder, watermark_path, output_folder, opacity=0.5):
    if not os.path.exists(output_folder):
        os.makedirs(output_folder)
    
    for filename in os.listdir(input_folder):
        if filename.endswith('.jpg'):
            image_path = os.path.join(input_folder, filename)
            output_path = os.path.join(output_folder, filename)
            add_watermark(image_path, watermark_path, output_path, opacity)

# 변수 설정
input_folder = 'sea_image'
watermark_path = 'chatgpt_logo.png'
output_folder = 'output_images'

# 이미지 처리
process_images(input_folder, watermark_path, output_folder, opacity=0.5)

!pip install rembg

import os
from rembg import remove
from PIL import Image

def remove_background(input_path, output_path):
    input_image = Image.open(input_path).convert('RGBA')
    output_image = remove(input_image)
    
    # RGBA 이미지를 PNG 형식으로 저장
    output_image.save(output_path, 'PNG')

def process_images(input_folder, output_folder):
    if not os.path.exists(output_folder):
        os.makedirs(output_folder)
    
    for filename in os.listdir(input_folder):
        if filename.endswith('.jpg'):
            input_path = os.path.join(input_folder, filename)
            output_path = os.path.join(output_folder, os.path.splitext(filename)[0] + '.png')  # PNG 확장자로 저장
            remove_background(input_path, output_path)

# 변수 설정
input_folder = 'people'
output_folder = 'people_no_bg'

# 이미지 처리
process_images(input_folder, output_folder)