Upload 2 files

app.py

@@ -0,0 +1,108 @@
+import gradio as gr
+import numpy as np
+from scipy.ndimage import uniform_filter, median_filter, gaussian_filter1d
+from skimage import exposure
+import matplotlib.pyplot as plt
+import io
+from PIL import Image
+
+def splitnormalize_balanced(img):
+    p_lo, p_hi = np.percentile(img, [2, 98])
+    img = np.clip((img - p_lo) / (p_hi - p_lo + 1e-6), 0, 1)
+    img = exposure.equalize_adapthist(img, clip_limit=0.003, nbins=256)
+    return img
+
+def removebias(img, regionwidth=10, fraction=0.75):
+    kz = 2*regionwidth + 1
+    kx = 2*regionwidth + 1
+    bg = uniform_filter(img, size=(kz, kx))
+    out = img - fraction * bg
+    out = (out - out.min()) / (out.max() - out.min() + 1e-6)
+    return out
+
+def estimate_medline_intensity(img):
+    H, W = img.shape
+    p1, p99 = np.percentile(img, [1, 99])
+    norm = np.clip((img - p1) / (p99 - p1 + 1e-6), 0, 1)
+    z_low = int(0.30 * H)
+    z_high = int(0.70 * H)
+    return np.argmax(norm[z_low:z_high, :], axis=0) + z_low
+
+def detect_rpe_simple(img, medline):
+    H, W = img.shape
+    sm = gaussian_filter1d(img, sigma=4, axis=0)
+    grad = np.gradient(gaussian_filter1d(sm, sigma=1, axis=0), axis=0)
+    rpe = np.argmax(-grad[int(0.30*H):int(0.85*H), :], axis=0) + int(0.30*H)
+    return rpe
+
+def linesweeter(y):
+    from scipy.signal import savgol_filter
+    return savgol_filter(y, 9, 2)
+
+def overlay(img, curve, title=""):
+    H, W = img.shape
+    x = np.arange(W)
+    fig, ax = plt.subplots(figsize=(8,4))
+    ax.imshow(img, cmap="gray")
+    ax.plot(x, curve, 'r-', lw=2)
+    ax.set_title(title)
+    ax.axis('off')
+    buf = io.BytesIO()
+    plt.savefig(buf, format="png", bbox_inches="tight")
+    buf.seek(0)
+    return Image.open(buf)
+
+with gr.Blocks(title="OCT Step-by-Step Visual Lab") as demo:
+    gr.Markdown("## 🧠 OCT Step-by-Step Visual Lab — Compare cada etapa lado a lado")
+
+    img_state = gr.State()
+
+    with gr.Tab("1) Carregar Imagem"):
+        img_input = gr.Image(label="Imagem OCT", type="numpy")
+        def store(img):
+            return img
+        img_input.change(store, inputs=img_input, outputs=img_state)
+
+    with gr.Tab("2) Normalização"):
+        btn_norm = gr.Button("Aplicar splitnormalize (balanced)")
+        out_norm = gr.Image(label="Resultado")
+        btn_norm.click(splitnormalize_balanced, inputs=img_state, outputs=out_norm)
+
+    with gr.Tab("3) Remove Bias"):
+        btn_bias = gr.Button("Aplicar removebias")
+        out_bias = gr.Image(label="Resultado")
+        btn_bias.click(removebias, inputs=img_state, outputs=out_bias)
+
+    with gr.Tab("4) Mediana 5×9"):
+        btn_med = gr.Button("Aplicar filtro mediano (5x9)")
+        out_med = gr.Image(label="Resultado")
+        btn_med.click(lambda i: median_filter(i, size=(5,9)), inputs=img_state, outputs=out_med)
+
+    with gr.Tab("5) Estimar Medline (IS/OS)"):
+        btn_medline = gr.Button("Calcular Medline")
+        out_medline = gr.Image(label="Visualização")
+        def show_medline(img):
+            med = estimate_medline_intensity(img)
+            return overlay(img, med, "Medline (IS/OS)")
+        btn_medline.click(show_medline, inputs=img_state, outputs=out_medline)
+
+    with gr.Tab("6) Detectar RPE"):
+        btn_rpe = gr.Button("Detectar RPE simples")
+        out_rpe = gr.Image(label="Visualização")
+        def step_rpe(img):
+            med = estimate_medline_intensity(img)
+            rpe = detect_rpe_simple(img, med)
+            return overlay(img, rpe, "RPE simples")
+        btn_rpe.click(step_rpe, inputs=img_state, outputs=out_rpe)
+
+    with gr.Tab("7) Suavização da RPE"):
+        btn_smooth = gr.Button("Suavizar RPE (linesweeter)")
+        out_smooth = gr.Image(label="Visualização")
+        def step_smooth(img):
+            med = estimate_medline_intensity(img)
+            rpe = detect_rpe_simple(img, med)
+            rpe_s = linesweeter(rpe)
+            return overlay(img, rpe_s, "RPE suavizada (linesweeter)")
+        btn_smooth.click(step_smooth, inputs=img_state, outputs=out_smooth)
+
+demo.launch()

requirements.txt

@@ -0,0 +1,5 @@
+gradio
+numpy
+scipy
+scikit-image
+matplotlib