Upload New File

api.py

+from flask import Flask, request
+from flask_cors import CORS
+import json
+import werkzeug
+import cv2
+from tensorflow.keras.models import load_model
+import numpy as np
+from rome_numeric_rec.ocr import get_rome_number
+from rome_numeric_rec.rome_to_number import roman_to_int
+import base64
+from imageio import imread
+import io
+from equation_solver.es import get_ans_from_image
+from numeric_pattern_rec.LinearRegression import get_next_numbers_of_sequence
+import numeric_pattern_rec.ocr as ocr
+
+from keras.models import Sequential
+from keras.layers import Dense, Activation, Flatten, Dropout
+from keras.layers import Conv2D, MaxPooling2D
+
+app = Flask(__name__)
+CORS(app)
+
+model = load_model('shape_rec/model.h5')
+
+data_dic = {0: 'rectangle', 1: 'circle', 2: 'ellipse', 3: 'hexagon', 4: 'trangle', 5: 'square', 6: 'stars'}
+
+
+@app.route('/equation_solver', methods=['GET', 'POST'])
+def equation_solver():
+    # for base 64 json input
+    string_64 = request.json['image']
+    string_64 = str(string_64).replace('image/webp;base64,', '')
+    img = base_64_cv2(string_64)
+
+    img_gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
+
+    answer = get_ans_from_image(img_gray)
+
+    return_str = '{ "answer" : "' + str(answer) + '" }'
+
+    print(return_str)
+
+    return json.loads(return_str)
+
+
+@app.route('/number_pattern', methods=['GET', 'POST'])
+def number_pattern():
+    string_64 = request.json['image']
+    string_64 = str(string_64).replace('image/webp;base64,', '')
+    img = base_64_cv2(string_64)
+    cv2.imwrite('upload/number_pattern.png', img)
+    n = request.json['n']
+
+    input_pattern = ocr.get_ocr('upload/number_pattern.png')
+    input_list = []
+
+    for i in input_pattern.split(','):
+
+        try:
+            input_list.append(int(str(i).replace(',', '').replace(' ', '')))
+        except:
+            pass
+
+    y, img = get_next_numbers_of_sequence(5, input_list)
+
+    base64_string = get_base_64(img)
+
+    return_str = '{ "predict_values" : ' + str(y) + ', "plot" : "' + str(base64_string) + '" }'
+
+    print(return_str)
+
+    return json.loads(return_str)
+
+
+@app.route('/rome_to_number', methods=['GET', 'POST'])
+def rome_to_number():
+    string_64 = request.json['image']
+    string_64 = str(string_64).replace('image/webp;base64,', '')
+    img = base_64_cv2(string_64)
+    cv2.imwrite('upload/number_pattern.png', img)
+
+    rome_text = get_rome_number(img).upper().replace(' ', '').replace('\n', '')
+    print(rome_text)
+    numeric_number = roman_to_int(rome_text)
+
+    return_str = '{ "res" : ' + str(numeric_number) + ' }'
+
+    print(return_str)
+
+    return json.loads(return_str)
+
+
+@app.route('/rome_to_number_new', methods=['GET', 'POST'])
+def rome_to_number_new():
+    string_64 = request.json['image']
+    string_64 = str(string_64).replace('image/webp;base64,', '')
+    img = base_64_cv2(string_64)
+    cv2.imwrite('upload/number_pattern.png', img)
+
+    rome_data_dic = {0: 'VI', 1: 'II', 2: 'I', 3: 'VII', 4: 'VIII', 5: 'IV', 6: 'IX', 7: 'V', 8: 'X', 9: 'III'}
+
+    data = np.load('rome_numeric_rec/Processed_Data/data/data.npy')
+    target = np.load('rome_numeric_rec/Processed_Data/data/target.npy')
+
+    model = Sequential()
+
+    model.add(Conv2D(256, (3, 3), input_shape=data.shape[1:]))
+    model.add(Activation('relu'))
+    model.add(MaxPooling2D(pool_size=(2, 2)))
+
+    model.add(Conv2D(128, (3, 3)))
+    model.add(Activation('relu'))
+    model.add(MaxPooling2D(pool_size=(2, 2)))
+
+    model.add(Flatten())
+
+    model.add(Dropout(0.5))
+    model.add(Dense(64, activation='relu'))
+
+    model.add(Dense(4, activation='softmax'))
+
+    model.compile(loss='categorical_crossentropy', optimizer='adam', metrics=['accuracy'])
+
+    model.load_weights('rome_numeric_rec/model.h5')
+
+    img = cv2.imread('upload/number_pattern.png', img)
+    gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
+    resized = cv2.resize(gray, (50, 50))
+    normalized = resized / 255.0
+    reshaped = np.reshape(normalized, (1, 50, 50, 1))
+
+    result = model.predict(reshaped)
+    label = np.argmax(result, axis=1)[0]
+    prob = np.max(result, axis=1)[0]
+    prob = round(prob, 2) * 100
+
+    print('Confident : ', np.max(result, axis=1)[0])
+
+    numeric_number = roman_to_int(rome_data_dic[np.argmax(result, axis=1)[0]])
+
+    return_str = '{ "res" : ' + str(numeric_number) + ' }'
+
+    print(return_str)
+
+    return json.loads(return_str)
+
+
+@app.route('/shape_classifier', methods=['GET', 'POST'])
+def shape_classifier():
+    string_64 = request.json['image']
+    string_64 = str(string_64).replace('image/webp;base64,', '')
+    img = base_64_cv2(string_64)
+    cv2.imwrite('upload/number_pattern.png', img)
+
+    img = np.asarray(img)
+    img = cv2.resize(img, (32, 32))
+    img = preProcessing(img)
+    img = img.reshape(1, 32, 32, 1)
+    classIndex = int(model.predict_classes(img))
+    predictions = model.predict(img)
+    # probVal = np.amax(predictions)
+    a = np.amax(predictions) * 100
+    x = "%.2f" % round(a, 2)
+    print(classIndex)
+    print(x)
+
+    return_str = '{ "score" : "' + str(data_dic[classIndex]) + '" }'
+
+    print(return_str)
+
+    return json.loads(return_str)
+
+
+def preProcessing(img):
+    img = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
+    img = cv2.equalizeHist(img)
+    img = img / 255
+    return img
+
+
+def base_64_cv2(string_64):
+    img = imread(io.BytesIO(base64.b64decode(string_64)))
+    cv2_img = cv2.cvtColor(img, cv2.COLOR_RGB2BGR)
+    return cv2_img
+
+
+def get_base_64(img):
+    retval, buffer = cv2.imencode('.jpg', img)
+    jpg_as_text = base64.b64encode(buffer)
+    return jpg_as_text
+
+
+if __name__ == '__main__':
+    app.run(host="0.0.0.0", port=5500, debug=True)