Iterating predictions even when no shark is sighted. Improving messages for info

app.py

@@ -60,8 +60,13 @@ def overlay_text_on_image(image, text_list, font=cv2.FONT_HERSHEY_SIMPLEX, font_
     y0, dy = margin, int(relative*0.1) # start y position and line gap
     for i, line in enumerate(text_list):
         y = y0 + i * dy
-        text_width, _ = cv2.getTextSize(line, font, font_size, font_thickness)[0]
-        cv2.putText(image, line, (image.shape[1] - text_width - margin, y), font, font_size, color, font_thickness, lineType=cv2.LINE_AA)
+        if 'Shark' in line or 'Human' in line :
+            text_width, _ = cv2.getTextSize(line, font, font_size*1.5, font_thickness)[0]
+            cv2.putText(image, line, (image.shape[1] - text_width - margin, y), font, font_size*1.2, color, font_thickness, lineType=cv2.LINE_AA)
+        else:
+            text_width, _ = cv2.getTextSize(line, font, font_size, font_thickness)[0]
+            cv2.putText(image, line, (image.shape[1] - text_width - margin, y), font, font_size, color, font_thickness, lineType=cv2.LINE_AA)
+            
     return image
 
 def overlay_logo(frame,logo, position=(10, 10)):
@@ -108,20 +113,46 @@ def draw_cockpit(frame, top_pred,cnt):
     # Bullet points:
     high_danger_color = (255,0,0)
     low_danger_color = yellowgreen = (154,205,50)
-    shark_sighted = 'Shark Detected: ' + str(top_pred['shark_sighted'])
-    human_sighted = 'Number of Humans: ' + str(top_pred['human_n'])
-    shark_size_estimate = 'Biggest shark size: ' + str(top_pred['biggest_shark_size'])
-    shark_weight_estimate = 'Biggest shark weight: ' + str(top_pred['biggest_shark_weight'])
+    
+    if top_pred['shark_sighted'] > 0:
+        shark_suspected = 'Shark Sighted !'
+    elif top_pred['shark_suspected'] > 0:
+        shark_suspected = 'Shark Suspected !'
+    else:
+        shark_suspected = 'No Sharks ...'
+
+    if top_pred['human_sighted'] > 0:
+        human_suspected = 'Human Sighted !'
+    elif top_pred['human_suspected'] > 0:
+        human_suspected = 'Human Suspected !'
+    else:
+        human_suspected = 'No Humans ...'
+
+    shark_size_estimate = 'Biggest shark size: ' + str(top_pred['biggest_shark_size']) if top_pred['biggest_shark_size'] else 'Biggest shark size: ...'
+    shark_weight_estimate = 'Biggest shark weight: ' + str(top_pred['biggest_shark_weight']) if top_pred['biggest_shark_weight'] else 'Biggest shark weight: ...'
+
     danger_level = 'Danger Level: ' 
-    danger_level += 'High' if top_pred['dangerous_dist'] else 'Low'
-    danger_color = 'orangered' if top_pred['dangerous_dist'] else 'yellowgreen'
+    danger_level += 'High' if top_pred['dangerous_dist_confirmed'] else 'Low'
+
+    danger_color = 'orangered' if top_pred['dangerous_dist_confirmed'] else 'yellowgreen'
+
     # Create a list of strings to plot
-    strings = [shark_sighted, human_sighted, shark_size_estimate, shark_weight_estimate, danger_level]
+    strings = [shark_suspected, human_suspected, shark_size_estimate, shark_weight_estimate, danger_level]
+    
+    # shark_sighted = 'Shark Detected: ' + str(top_pred['shark_sighted'])
+    # human_sighted = 'Number of Humans: ' + str(top_pred['human_n'])
+    # shark_size_estimate = 'Biggest shark size: ' + str(top_pred['biggest_shark_size'])
+    # shark_weight_estimate = 'Biggest shark weight: ' + str(top_pred['biggest_shark_weight'])
+    # danger_level = 'Danger Level: ' 
+    # danger_level += 'High' if top_pred['dangerous_dist'] else 'Low'
+    # danger_color = 'orangered' if top_pred['dangerous_dist'] else 'yellowgreen'
+    # # Create a list of strings to plot
+    # strings = [shark_sighted, human_sighted, shark_size_estimate, shark_weight_estimate, danger_level]
     relative = max(frame.shape[0],frame.shape[1])
-    if top_pred['shark_sighted'] and top_pred['dangerous_dist'] and cnt%2 == 0:
+    if top_pred['shark_sighted'] and top_pred['dangerous_dist_confirmed'] and cnt%2 == 0:
         frame  = add_border(frame, color=high_danger_color, thickness=int(relative*0.025))
         frame = add_danger_symbol_from_image(frame, top_pred)
-    elif top_pred['shark_sighted'] and not top_pred['dangerous_dist'] and cnt%2 == 0:
+    elif top_pred['shark_sighted'] and not top_pred['dangerous_dist_confirmed'] and cnt%2 == 0:
          frame  = add_border(frame, color=low_danger_color, thickness=int(relative*0.025))
          frame = add_danger_symbol_from_image(frame, top_pred)
     else: 
@@ -192,13 +223,14 @@ def process_video(input_video, out_fps = 'auto', skip_frames = 7):
             #
             #video.write(cv2.cvtColor(frame, cv2.COLOR_RGB2BGR))
             
-            if cnt*skip_frames %2==0 and top_pred['shark_sighted']: 
+            if cnt*skip_frames %2==0: 
                 prediction_frame = cv2.resize(prediction_frame, (int(width), int(height)))
                 frame = prediction_frame
                 
-            if  top_pred['shark_sighted']: 
+            #if  top_pred['shark_sighted'] or top_pred['shark_suspected']: 
                 frame = draw_cockpit(frame, top_pred,cnt*skip_frames)
             
+            
             frame = cv2.resize(frame, (int(width), int(height)))
             video.write(cv2.cvtColor(frame, cv2.COLOR_RGB2BGR))
                 
@@ -208,10 +240,10 @@ def process_video(input_video, out_fps = 'auto', skip_frames = 7):
             drawn_count += 1
             #print('sending frame')
             print('finalizing frame:',cnt)
-            print(pred_dashbord.shape)
-            print(frame.shape)
-            print(prediction_frame.shape)
-            print(width, height)
+            #print(pred_dashbord.shape)
+            #print(frame.shape)
+            #print(prediction_frame.shape)
+            #print(width, height)
             yield frame , None
         
         cnt += 1

metrics.py

@@ -50,7 +50,7 @@ def mask_to_boxes(mask):
     return ltrb_boxes
 
 
-def get_top_predictions(prediction = None, threshold = 0.1):
+def get_top_predictions(prediction = None, threshold = 0.8):
     if prediction is None:
         return None, None
     else:
@@ -68,10 +68,12 @@ def add_class_labels(top_pred = {}, class_labels = None):
         return top_pred
     else:
         top_pred['pred_above_thresh_labels'] = [class_labels[x].lower() for x in top_pred['pred_above_thresh']]
+        #print(top_pred['pred_above_thresh_labels'])
+        #print(top_pred['pred_above_thresh_scores'])
         top_pred['any_detection'] = len(top_pred['pred_above_thresh_labels']) > 0
         if top_pred['any_detection']:
             # Get shark / human / unknown vectors
-            top_pred['is_shark'] = np.array([1 if 'shark' in x else 0 for x in top_pred['pred_above_thresh_labels']])
+            top_pred['is_shark'] = np.array([1 if 'shark' in x and 'shadow' not in x else 0 for x in top_pred['pred_above_thresh_labels']])
             top_pred['is_human'] = np.array([1 if 'person' in x else 1 if 'surfer' in x else 0 for x in top_pred['pred_above_thresh_labels']])
             top_pred['is_unknown'] = np.array([1 if 'unidentifiable' in x else 0 for x in top_pred['pred_above_thresh_labels']])
             # Get shark / human / unknown  numbers of detections
@@ -101,7 +103,7 @@ def add_class_sizes(top_pred = {}, class_sizes = None):
                 else:
                     size_list.append(tmp_class_sizes['feet'])
 
-                if 'shark' in tmp_pred.lower():
+                if 'shark' in tmp_pred.lower() and 'shadow' not in tmp_pred.lower() :
                     shark_size_list.append(np.mean(tmp_class_sizes['feet']))
 
         top_pred['pred_above_thresh_sizes'] = size_list