BACKEND :errors fixed in methods

backend/app.py

@@ -55,7 +55,7 @@ def create_tables():
 
 
 @app.get('/api/v1/')
-def index():
+async def index():
     return 'UML Diagram Plagiarism Detection Tool API'
 
 

backend/routes/submission_routes.py

@@ -14,7 +14,7 @@ submission = Blueprint('submissions', __name__, url_prefix='/api/v1/submissions'
 
 @submission.post('/upload')
 @jwt_required()
-def upload_submission():
+async def upload_submission():
     user_id = get_jwt_identity()
 
     image = request.files['file']
@@ -29,9 +29,10 @@ def upload_submission():
         use_case_obj = save_submission(assignment_id, image, submission_type, comment, user_id)
         model_object_detection(image.filename, use_case_obj.id)
         return jsonify({'filename': image.filename}), HTTP_200_OK
+
     elif submission_type == 'class':
         class_obj = save_submission(assignment_id, image, submission_type, comment, user_id)
-        component_separation(image.filename, class_obj.id)
+        await component_separation(image.filename, class_obj.id)
         return jsonify({'id': str(class_obj.id)}), HTTP_200_OK
     else:
         return jsonify({'err': 'invalid request '}), HTTP_400_BAD_REQUEST

backend/services/class_model_detection_service.py

 import operator
+import os
 import re
 
 import cv2
@@ -12,30 +13,37 @@ import app
 import tensorflow as tf
 import spacy
 
+
 from config.database import db
 from models.method import Method
 
+ts.pytesseract.tesseract_cmd = r'C:\Users\DELL\AppData\Local\Programs\Tesseract-OCR\tesseract.exe'
+
 
 def component_separation(filename, class_comp_id):
     mdl1_path = app.CLASS_SAVED_MODEL_PATH
     lbl1_path = app.CLASS_SAVED_LABEL_PATH
     img1_path = app.SUBMISSION_PATH_CLASS + '/' + filename
+    image_nparray = np.array(Image.open(img1_path))
+
+    print(img1_path)
     boxes, num_entities, accurate_indexes, num_entities, category_index, class_id = class_object_detection(mdl1_path,
                                                                                                            lbl1_path,
-                                                                                                           img1_path)
+                                                                                                           image_nparray)
 
     # Convert the class id in their name
     if num_entities > 0:
         for index in range(0, len(accurate_indexes)):
-
-            if category_index[class_id[index]]['name'] == 'class' or category_index[class_id]['name'] == 'interface':
-                class_details_detection(filename, boxes, index, class_comp_id)
+            if category_index[class_id[index]]['name'] == 'class':
+                print(filename)
+                _image = crop_image_(image_nparray, boxes, index)
+                class_details_detection(_image, class_comp_id)
 
 
-def class_object_detection(model_path, label_path, image_path):
+def class_object_detection(model_path, label_path, image_nparray):
     detect_fn = tf.saved_model.load(model_path)
     category_index = label_map_util.create_category_index_from_labelmap(label_path, use_display_name=True)
-    image_np = np.array(Image.open(image_path))
+    image_np = image_nparray
     input_tensor = tf.convert_to_tensor(image_np)
 
     input_tensor = input_tensor[tf.newaxis, ...]
@@ -57,31 +65,34 @@ def class_object_detection(model_path, label_path, image_path):
     return boxes, num_entities, accurate_indexes, num_entities, category_index, class_id
 
 
-def class_details_detection(filename, boxes, index, class_comp_id):
-    _image = crop_and_image_resolution(filename, boxes, index)
-    cv2.imwrite(app.SUBMISSION_PATH_CLASS + '/' + "class" + str(index), _image)
+def class_details_detection(_image, class_comp_id):
+    retval = os.getcwd()
+    print("Current working directory %s" % retval)
+    # _image.save(os.path.join(app.SUBMISSION_PATH_CLASS, "class" + str(index) + ".jpg"))
     mdl2_path = app.CLASS_COMP_SAVED_MODEL_PATH
     lbl2_path = app.CLASS_COMP_SAVED_LABEL_PATH
-    img2_path = app.SUBMISSION_PATH_CLASS + '/' + "class" + str(index)
+    # img2_path = app.SUBMISSION_PATH_CLASS + '/' + "class" + str(index)
     boxes_class, num_entities, accurate_indexes, num_entities, category_index, class_id = class_object_detection(
-        mdl2_path, lbl2_path, img2_path)
+        mdl2_path, lbl2_path, _image)
 
     if num_entities > 1:
         for j in range(0, len(accurate_indexes)):
             if category_index[class_id[j]]['name'] == 'class_attributes':
-                class_attributes = crop_and_image_resolution(img2_path, boxes_class, j)
+                class_attributes = crop_image_(_image, boxes_class, j)
                 text = text_extraction(class_attributes)
+                print(text)
                 save_attribute_method(text, 'attribute')
             elif category_index[class_id[j]]['name'] == 'class_methods':
-                class_methods = crop_and_image_resolution(img2_path, boxes_class, j)
+                class_methods = crop_image_(_image, boxes_class, j)
                 text = text_extraction(class_methods)
+                print(text)
                 save_attribute_method(text, 'method')
 
         # class name detection and save the class object
 
 
-def crop_and_image_resolution(path, boxes, index):
-    image = cv2.imread(path)
+def crop_image_(image, boxes, index):
+    # image = cv2.imread(path)
     height, width, c = image.shape
     # crop box format: xmin, ymin, xmax, ymax
     ymin = boxes[index][0] * height
@@ -90,10 +101,10 @@ def crop_and_image_resolution(path, boxes, index):
     xmax = boxes[index][3] * width
 
     cropped_image = image[int(ymin):int(ymax), int(xmin):int(xmax)]
-    image = cv2.cvtColor(cropped_image, cv2.COLOR_BGR2GRAY)
-    image = cv2.resize(image, (800, 500))
+    # image = cv2.cvtColor(cropped_image, cv2.COLOR_BGR2GRAY)
+    # image = cv2.resize(image, (800, 500))
 
-    return image
+    return cropped_image
 
 
 def text_extraction(image):