Update app.py

app.py

@@ -2,44 +2,47 @@ import gradio as gr
 import cv2
 import numpy as np
 
-def apply_filter(image, filter_type):
+def apply_filter(image, filter_type, intensity):
+    # 강도를 0.0에서 1.0 사이로 정규화
+    normalized_intensity = intensity / 100.0
+
     if filter_type == "Grayscale":
         # 흑백 변환
         return convert_to_grayscale(image)
     elif filter_type == "Soft Glow":
         # Soft Glow 효과
         gaussian = cv2.GaussianBlur(image, (15, 15), 0)
-        soft_glow = cv2.addWeighted(image, 0.5, gaussian, 0.5, 0)
+        soft_glow = cv2.addWeighted(image, 1 - normalized_intensity, gaussian, normalized_intensity, 0)
         return soft_glow
     elif filter_type == "Portrait Enhancer":
         # 간단한 포트레이트 향상 (부드럽게 조정된 대비 및 밝기)
-        enhanced = cv2.detailEnhance(image, sigma_s=5, sigma_r=0.1)
+        enhanced = cv2.detailEnhance(image, sigma_s=5, sigma_r=0.1 + 0.15 * normalized_intensity)
         return enhanced
     elif filter_type == "Warm Tone":
         # 따뜻한 톤 효과 (빨강 및 노랑 톤을 강조)
-        increase_red = np.array([[1.2, 0.0, 0.0],
+        increase_red = np.array([[1.0 + 0.2 * normalized_intensity, 0.0, 0.0],
                                  [0.0, 1.0, 0.0],
-                                 [0.0, 0.0, 0.8]])
+                                 [0.0, 0.0, 1.0 - 0.2 * normalized_intensity]])
         warm_image = cv2.transform(image, increase_red)
         return warm_image
     elif filter_type == "Cold Tone":
         # 차가운 톤 효과 (파랑 톤을 강조)
-        increase_blue = np.array([[0.8, 0.0, 0.0],
+        increase_blue = np.array([[1.0 - 0.2 * normalized_intensity, 0.0, 0.0],
                                   [0.0, 1.0, 0.0],
-                                  [0.0, 0.0, 1.2]])
+                                  [0.0, 0.0, 1.0 + 0.2 * normalized_intensity]])
         cold_image = cv2.transform(image, increase_blue)
         return cold_image
     elif filter_type == "High-Key":
         # High-Key 효과 (밝기 증가)
-        high_key = cv2.convertScaleAbs(image, alpha=1.2, beta=30)
+        high_key = cv2.convertScaleAbs(image, alpha=1.0 + 0.2 * normalized_intensity, beta=30)
         return high_key
     elif filter_type == "Low-Key":
         # Low-Key 효과 (밝기 감소)
-        low_key = cv2.convertScaleAbs(image, alpha=0.7, beta=-30)
+        low_key = cv2.convertScaleAbs(image, alpha=1.0 - 0.3 * normalized_intensity, beta=-30)
         return low_key
     elif filter_type == "Haze":
         # Haze 효과
-        haze = cv2.addWeighted(image, 0.7, np.full(image.shape, 255, dtype=np.uint8), 0.3, 0)
+        haze = cv2.addWeighted(image, 1.0 - 0.3 * normalized_intensity, np.full(image.shape, 255, dtype=np.uint8), 0.3 * normalized_intensity, 0)
         return haze
     else:
         return image
@@ -49,9 +52,9 @@ def convert_to_grayscale(image):
     gray_image = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)
     return gray_image
 
-def convert_and_save(image, filter_type):
-    # 선택한 필터를 이미지에 적용
-    filtered_image = apply_filter(image, filter_type)
+def convert_and_save(image, filter_type, intensity):
+    # 선택한 필터와 강도를 이미지에 적용
+    filtered_image = apply_filter(image, filter_type, intensity)
     # 이미지를 저장
     output_path = "output.jpg"
     cv2.imwrite(output_path, filtered_image)
@@ -65,11 +68,12 @@ iface = gr.Interface(
         gr.Radio(
             ["Grayscale", "Soft Glow", "Portrait Enhancer", "Warm Tone", "Cold Tone", "High-Key", "Low-Key", "Haze"], 
             label="필터 선택"
-        )
+        ),
+        gr.Slider(minimum=1, maximum=100, default=50, label="필터 강도")
     ], 
     outputs=["image", "file"], 
     title="이미지 필터 및 흑백 변환기",
-    description="이미지를 업로드하고 필터(또는 흑백 변환)를 선택하면, 변환된 이미지를 JPG 파일로 다운로드할 수 있습니다."
+    description="이미지를 업로드하고 필터와 강도를 선택하면, 변환된 이미지를 JPG 파일로 다운로드할 수 있습니다."
 )
 
 iface.launch()