Add KDE func

web_app.py

@@ -25,7 +25,7 @@ DETECTION_MODEL_x = os.path.join(DIR_NAME, 'models', 'YOLOv8-X_CNO_Detection.pt'
 # cno_df = pd.DataFrame()
 
 
-def predict_image(name, model, img, conf_threshold, iou_threshold):
+def predict_image(name, img_h, img_w, model, img, conf_threshold, iou_threshold):
     """Predicts and plots labeled objects in an image using YOLOv8 model with adjustable confidence and IOU thresholds."""
     gr.Info("Starting process")
     # gr.Warning("Name is empty")
@@ -59,29 +59,31 @@ def predict_image(name, model, img, conf_threshold, iou_threshold):
     cno_image = []
     kde_image = []
     file_name = []
+    ecti_score = []
 
-    total_layer_area = []
-    total_layer_cno = []
-    total_layer_density = []
-    avg_area_col = []
-    total_area_col = []
+    # total_layer_area = []
+    # total_layer_cno = []
+    # total_layer_density = []
+    # avg_area_col = []
+    # total_area_col = []
 
     for idx, result in enumerate(results):
         cno = len(result.boxes)
 
         file_label = img[idx].split(os.sep)[-1]
-        single_layer_area = []
-        single_layer_cno = []
+        # single_layer_area = []
+        # single_layer_cno = []
         single_layer_density = []
         total_area = 0
         if cno < 5:
-            avg_area_col.append(np.nan)
-            total_area_col.append(np.nan)
-            nan_arr = np.empty([25])
-            nan_arr[:] = np.nan
-            total_layer_area.append(nan_arr)
-            total_layer_cno.append(nan_arr)
-            total_layer_density.append(nan_arr)
+            # avg_area_col.append(np.nan)
+            # total_area_col.append(np.nan)
+            # nan_arr = np.empty([25])
+            # nan_arr[:] = np.nan
+            ecti_score.append(np.nan)
+            # total_layer_area.append(nan_arr)
+            # total_layer_cno.append(nan_arr)
+            # total_layer_density.append(nan_arr)
         else:
             cno_coor = np.empty([cno, 2], dtype=int)
 
@@ -125,8 +127,7 @@ def predict_image(name, model, img, conf_threshold, iou_threshold):
                                                                                                       (tf - ti)))
             kde.bandwidth = bw
             _ = kde.fit(cno_coor)
-            print("deb", result.orig_img.shape[1])
-            print("deb", result.orig_img.shape[0])
+
             xgrid = np.arange(0, result.orig_img.shape[1], 1)
             ygrid = np.arange(0, result.orig_img.shape[0], 1)
             xv, yv = np.meshgrid(xgrid, ygrid)
@@ -154,15 +155,18 @@ def predict_image(name, model, img, conf_threshold, iou_threshold):
                 if layer_area == 0:
                     density = np.round(0.0, 4)
                 else:
-                    density = np.round((ecno / layer_area) * result.orig_img.shape[0] * result.orig_img.shape[1] / 400, 4)
+                    density = np.round((ecno / layer_area) * result.orig_img.shape[0] * result.orig_img.shape[1] / (img_h * img_w), 4)
                 print("Level {}: Area={}, CNO={}, density={}".format(j, layer_area, ecno, density))
-                single_layer_area.append(layer_area)
-                single_layer_cno.append(ecno)
+                # single_layer_area.append(layer_area)
+                # single_layer_cno.append(ecno)
                 single_layer_density.append(density)
 
-            total_layer_area.append(single_layer_area)
-            total_layer_cno.append(single_layer_cno)
-            total_layer_density.append(single_layer_density)
+            # total_layer_area.append(single_layer_area)
+            # total_layer_cno.append(single_layer_cno)
+            # total_layer_density.append(single_layer_density)
+            # print(sum_range(single_layer_density, 10, 14))
+            # print("deb ", single_layer_density)
+            ecti_score.append(np.round(sum_range(single_layer_density, 10, 14) / 5.0, 2))
 
             
             # Plot CNO Distribution
@@ -174,30 +178,19 @@ def predict_image(name, model, img, conf_threshold, iou_threshold):
             plt.xlim(0, gdim[1] - 1)
             plt.ylim(gdim[0] - 1, 0)
             plt.plot()
-            # plt.show()
 
-            # plt.savefig("test.png", format='png', bbox_inches='tight', pad_inches=0)
-            # plt.figure()
-            # plt.plot([1, 2])
             img_buf = io.BytesIO()
             plt.savefig(img_buf, format='png', bbox_inches='tight', pad_inches=0)
             kde_im = Image.open(img_buf)
-            # kde_im.show()
-
-            # kde_img = Image.frombytes('RGB', fig.canvas.get_width_height(), fig.canvas.tostring_rgb())
-            # kde_image.append([imgplot, file_label])
             kde_image.append([kde_im, file_label])
-            #plt.savefig(os.path.join(kde_dir, '{}_{}_{}_KDE.png'.format(file_list[idx], model_type, conf)),
-            #            bbox_inches='tight', pad_inches=0)
-
-            #img_buf.close()
-        ### ============================
 
     data = {
         "Files": file_name,
         "CNO Count": cno_count,
+        "ECTI Score": ecti_score
     }
 
+
     # load data into a DataFrame object:
     cno_df = pd.DataFrame(data)
 
@@ -225,8 +218,11 @@ def highlight_df(df, data: gr.SelectData):
     # print("selected", data.value["caption"])
     return data.value["caption"], styler
 
+
 def reset():
     name_textbox = ""
+    img_h = 20
+    img_w = 20
     gender_radio = None
     age_slider = 0
     fitzpatrick = 1
@@ -240,16 +236,26 @@ def reset():
     cno_gallery = []
     test_label = ""
 
-    return name_textbox, gender_radio, age_slider, fitzpatrick, history, model_radio, input_files, conf_slider, \
+    return name_textbox, img_h, img_w, gender_radio, age_slider, fitzpatrick, history, model_radio, input_files, conf_slider, \
         iou_slider, analysis_results, afm_gallery, cno_gallery, test_label
 
+def sum_range(l,a,b):
+    s = 0
+    for i in range(a,b+1):
+        s += l[i]
+    return s
+
 
 with gr.Blocks(title="AFM AI Analysis", theme="default") as app:
     with gr.Row():
         with gr.Column():
             # gr.Markdown("User Information")
             with gr.Accordion("User Information", open=True):
-                name_textbox = gr.Textbox(label="Name")
+                with gr.Row():
+                    name_textbox = gr.Textbox(label="Sample")
+                    with gr.Row():
+                        img_h = gr.Number(label="Image Height (μm)", value=20, interactive=True)
+                        img_w = gr.Number(label="Image Width (μm)", value=20, interactive=True)
                 with gr.Row():
                     gender_radio = gr.Radio(["Male", "Female"], label="Gender", interactive=True, scale=1)
                     age_slider = gr.Slider(minimum=0, maximum=100, step=1, value=0, label="Age", interactive=True, scale=2)
@@ -280,11 +286,11 @@ with gr.Blocks(title="AFM AI Analysis", theme="default") as app:
 
     analyze_btn.click(
         fn=predict_image,
-        inputs=[name_textbox, model_radio, input_files, conf_slider, iou_slider],
+        inputs=[name_textbox, img_h, img_w, model_radio, input_files, conf_slider, iou_slider],
         outputs=[analysis_results, afm_gallery, cno_gallery, kde_gallery]
     )
 
-    clear_btn.click(reset, outputs=[name_textbox, gender_radio, age_slider, fitzpatrick, history, model_radio,
+    clear_btn.click(reset, outputs=[name_textbox, img_h, img_w, gender_radio, age_slider, fitzpatrick, history, model_radio,
                                     input_files, conf_slider, iou_slider, analysis_results, afm_gallery, cno_gallery,
                                     test_label])