license-plate-recognition-using-cnn (6).ipynb

{"metadata":{"kernelspec":{"language":"python","display_name":"Python 3","name":"python3"},"language_info":{"name":"python","version":"3.7.6","mimetype":"text/x-python","codemirror_mode":{"name":"ipython","version":3},"pygments_lexer":"ipython3","nbconvert_exporter":"python","file_extension":".py"},"kaggle":{"accelerator":"none","dataSources":[{"sourceId":1414839,"sourceType":"datasetVersion","datasetId":827967},{"sourceId":7044618,"sourceType":"datasetVersion","datasetId":4053559},{"sourceId":7044660,"sourceType":"datasetVersion","datasetId":4053584},{"sourceId":7044767,"sourceType":"datasetVersion","datasetId":4053659},{"sourceId":7044788,"sourceType":"datasetVersion","datasetId":4053673},{"sourceId":7045032,"sourceType":"datasetVersion","datasetId":4053845},{"sourceId":7046704,"sourceType":"datasetVersion","datasetId":4054954},{"sourceId":7048070,"sourceType":"datasetVersion","datasetId":4055835}],"dockerImageVersionId":29994,"isInternetEnabled":false,"language":"python","sourceType":"notebook","isGpuEnabled":false}},"nbformat_minor":4,"nbformat":4,"cells":[{"cell_type":"code","source":"import matplotlib.pyplot as plt\nimport numpy as np\nimport cv2\nimport tensorflow as tf\nfrom sklearn.metrics import f1_score \nfrom tensorflow.keras import optimizers\nfrom tensorflow.keras.models import Sequential\nfrom tensorflow.keras.preprocessing.image import ImageDataGenerator\nfrom tensorflow.keras.layers import Dense, Flatten, MaxPooling2D, Dropout, Conv2D","metadata":{"id":"ycRjhI25UC-P","execution":{"iopub.status.busy":"2023-11-25T05:42:23.887841Z","iopub.execute_input":"2023-11-25T05:42:23.888266Z","iopub.status.idle":"2023-11-25T05:42:32.261620Z","shell.execute_reply.started":"2023-11-25T05:42:23.888230Z","shell.execute_reply":"2023-11-25T05:42:32.260402Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"import cv2\nimport matplotlib.pyplot as plt\n\nimg = cv2.imread('/kaggle/input/punjabnumberplates/DSC_0346.jpg')\n\nheight, width, _ = img.shape\nleft_crop = int(width * 0.23)\nright_crop = int(width * 0.18)\n\ncropped_img = img[:, left_crop:width - right_crop]\n\ncv2.imwrite('cropped_image.png', cropped_img)\n\nplt.imshow(cv2.cvtColor(cropped_img, cv2.COLOR_BGR2RGB))\nplt.axis('off')\nplt.show()","metadata":{"execution":{"iopub.status.busy":"2023-11-25T07:03:46.622953Z","iopub.execute_input":"2023-11-25T07:03:46.623457Z","iopub.status.idle":"2023-11-25T07:03:46.758772Z","shell.execute_reply.started":"2023-11-25T07:03:46.623402Z","shell.execute_reply":"2023-11-25T07:03:46.757581Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"cropped_img = cv2.imread('cropped_image.png')\n\nheight, width, _ = cropped_img.shape\n\ndivision_point = int(height * 0.5)\n\n# Divide the image into two parts\nfirst_half = cropped_img[:division_point, :]\nsecond_half = cropped_img[division_point:, :]\n\nplt.subplot(1, 2, 1)\nplt.imshow(cv2.cvtColor(first_half, cv2.COLOR_BGR2RGB))\nplt.title('First Half')\nplt.axis('off')\n\nplt.subplot(1, 2, 2)\nplt.imshow(cv2.cvtColor(second_half, cv2.COLOR_BGR2RGB))\nplt.title('Second Half')\nplt.axis('off')\n\nplt.tight_layout()\n\nplt.show()","metadata":{"execution":{"iopub.status.busy":"2023-11-25T06:48:42.401972Z","iopub.execute_input":"2023-11-25T06:48:42.402372Z","iopub.status.idle":"2023-11-25T06:48:42.547541Z","shell.execute_reply.started":"2023-11-25T06:48:42.402336Z","shell.execute_reply":"2023-11-25T06:48:42.546233Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"def find_contours(dimensions, img) :\n    \n    cntrs, _ = cv2.findContours(img.copy(), cv2.RETR_TREE, cv2.CHAIN_APPROX_SIMPLE)\n\n    # Retrieve potential dimensions\n    lower_width = 50\n    upper_width = 100\n    lower_height = 10\n    upper_height = 70\n    \n    # Check largest 5 or  15 contours for license plate or character respectively\n    cntrs = sorted(cntrs, key=cv2.contourArea, reverse=True)[:15]\n    \n    ii = cv2.imread('contour.jpg')\n    \n    x_cntr_list = []\n    target_contours = []\n    img_res = []\n    for cntr in cntrs :\n        # detects contour in binary image and returns the coordinates of rectangle enclosing it\n        intX, intY, intWidth, intHeight = cv2.boundingRect(cntr)\n        \n        # checking the dimensions of the contour to filter out the characters by contour's size\n        if intWidth > lower_width and intWidth < upper_width and intHeight > lower_height and intHeight < upper_height :\n            x_cntr_list.append(intX) #stores the x coordinate of the character's contour, to used later for indexing the contours\n\n            char_copy = np.zeros((44,24))\n            # extracting each character using the enclosing rectangle's coordinates.\n            char = img[intY:intY+intHeight, intX:intX+intWidth]\n            char = cv2.resize(char, (20, 40))\n            \n            cv2.rectangle(ii, (intX,intY), (intWidth+intX, intY+intHeight), (50,21,200), 2)\n            plt.imshow(ii, cmap='gray')\n\n            # Make result formatted for classification: invert colors\n            char = cv2.subtract(255, char)\n\n            # Resize the image to 24x44 with black border\n            char_copy[2:42, 2:22] = char\n            char_copy[0:2, :] = 0\n            char_copy[:, 0:2] = 0\n            char_copy[42:44, :] = 0\n            char_copy[:, 22:24] = 0\n\n            img_res.append(char_copy) # List that stores the character's binary image (unsorted)\n            \n    # Return characters on ascending order with respect to the x-coordinate (most-left character first)\n            \n    plt.show()\n    # arbitrary function that stores sorted list of character indeces\n    indices = sorted(range(len(x_cntr_list)), key=lambda k: x_cntr_list[k])\n    img_res_copy = []\n    for idx in indices:\n        img_res_copy.append(img_res[idx])# stores character images according to their index\n    img_res = np.array(img_res_copy)\n\n    return img_res","metadata":{"id":"MzopHrMvUC-Z","execution":{"iopub.status.busy":"2023-11-25T06:48:44.991584Z","iopub.execute_input":"2023-11-25T06:48:44.992325Z","iopub.status.idle":"2023-11-25T06:48:45.012203Z","shell.execute_reply.started":"2023-11-25T06:48:44.992270Z","shell.execute_reply":"2023-11-25T06:48:45.010911Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"def segment_characters(image) :\n    img_lp = cv2.resize(image, (333, 75))\n    img_gray_lp = cv2.cvtColor(img_lp, cv2.COLOR_BGR2GRAY)\n    _, img_binary_lp = cv2.threshold(img_gray_lp, 200, 255, cv2.THRESH_BINARY+cv2.THRESH_OTSU)\n    img_binary_lp = cv2.erode(img_binary_lp, (3,3))\n    img_binary_lp = cv2.dilate(img_binary_lp, (3,3))\n\n    LP_WIDTH = img_binary_lp.shape[0]\n    LP_HEIGHT = img_binary_lp.shape[1]\n\n    img_binary_lp[0:3,:] = 255\n    img_binary_lp[:,0:3] = 255\n    img_binary_lp[72:75,:] = 255\n    img_binary_lp[:,330:333] = 255\n\n    dimensions = [LP_WIDTH/6,\n                       LP_WIDTH/3,\n                       LP_HEIGHT/3,\n                       2*LP_HEIGHT/2]\n    plt.imshow(img_binary_lp, cmap='gray')\n    plt.show()\n    cv2.imwrite('contour.jpg',img_binary_lp)\n    \n    char_list = find_contours(dimensions, img_binary_lp)\n\n    return char_list","metadata":{"id":"h23diSmEUC-e","execution":{"iopub.status.busy":"2023-11-25T06:48:48.731208Z","iopub.execute_input":"2023-11-25T06:48:48.731645Z","iopub.status.idle":"2023-11-25T06:48:48.745268Z","shell.execute_reply.started":"2023-11-25T06:48:48.731607Z","shell.execute_reply":"2023-11-25T06:48:48.743670Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"char_FirstHalf = segment_characters(first_half)","metadata":{"execution":{"iopub.status.busy":"2023-11-25T06:48:51.734254Z","iopub.execute_input":"2023-11-25T06:48:51.734911Z","iopub.status.idle":"2023-11-25T06:48:52.073892Z","shell.execute_reply.started":"2023-11-25T06:48:51.734842Z","shell.execute_reply":"2023-11-25T06:48:52.072578Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"char_SecondHalf = segment_characters(second_half)","metadata":{"id":"OGhFmSnYUC-j","outputId":"e9aea986-3422-4173-b4d8-11582f80d2c2","execution":{"iopub.status.busy":"2023-11-25T06:48:54.230064Z","iopub.execute_input":"2023-11-25T06:48:54.230510Z","iopub.status.idle":"2023-11-25T06:48:54.575016Z","shell.execute_reply.started":"2023-11-25T06:48:54.230473Z","shell.execute_reply":"2023-11-25T06:48:54.573673Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"for i in range(3):\n    plt.subplot(1, 10, i+1)\n    plt.imshow(char_FirstHalf[i], cmap='gray')\n    plt.axis('off')","metadata":{"execution":{"iopub.status.busy":"2023-11-25T06:48:56.350755Z","iopub.execute_input":"2023-11-25T06:48:56.351147Z","iopub.status.idle":"2023-11-25T06:48:56.481799Z","shell.execute_reply.started":"2023-11-25T06:48:56.351113Z","shell.execute_reply":"2023-11-25T06:48:56.480956Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"for i in range(4):\n    plt.subplot(1, 10, i+1)\n    plt.imshow(char_SecondHalf[i], cmap='gray')\n    plt.axis('off')","metadata":{"id":"rZoiyrDaUC-p","outputId":"e9210536-ff7e-4214-ce8a-ff1274656036","execution":{"iopub.status.busy":"2023-11-25T06:48:58.508802Z","iopub.execute_input":"2023-11-25T06:48:58.509477Z","iopub.status.idle":"2023-11-25T06:48:58.679994Z","shell.execute_reply.started":"2023-11-25T06:48:58.509405Z","shell.execute_reply":"2023-11-25T06:48:58.678753Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"### Model for characters","metadata":{"id":"QXhqHfXLUC-9"}},{"cell_type":"code","source":"import tensorflow.keras.backend as K\ntrain_datagen = ImageDataGenerator(rescale=1./255, width_shift_range=0.1, height_shift_range=0.1)\npath = '../input/ai-indian-license-plate-recognition-data/data/data'\ntrain_generator = train_datagen.flow_from_directory(\n        path+'/train',\n        target_size=(28,28),\n        batch_size=1,\n        class_mode='sparse')\n\nvalidation_generator = train_datagen.flow_from_directory(\n        path+'/val',\n        target_size=(28,28),\n        class_mode='sparse')","metadata":{"id":"BhrsmfX9UC_p","outputId":"55577343-eb1d-4980-99c0-59ecf8eefd99","execution":{"iopub.status.busy":"2023-11-25T06:49:02.013071Z","iopub.execute_input":"2023-11-25T06:49:02.013550Z","iopub.status.idle":"2023-11-25T06:49:02.290885Z","shell.execute_reply.started":"2023-11-25T06:49:02.013508Z","shell.execute_reply":"2023-11-25T06:49:02.289701Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"def f1score(y, y_pred):\n  return f1_score(y, tf.math.argmax(y_pred, axis=1), average='micro') \n\ndef custom_f1score(y, y_pred):\n  return tf.py_function(f1score, (y, y_pred), tf.double)","metadata":{"id":"WXdiO1Kq9kPI","execution":{"iopub.status.busy":"2023-11-25T06:49:07.318352Z","iopub.execute_input":"2023-11-25T06:49:07.318784Z","iopub.status.idle":"2023-11-25T06:49:07.326221Z","shell.execute_reply.started":"2023-11-25T06:49:07.318738Z","shell.execute_reply":"2023-11-25T06:49:07.324884Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"K.clear_session()\nmodel = Sequential()\nmodel.add(Conv2D(16, (22,22), input_shape=(28, 28, 3), activation='relu', padding='same'))\nmodel.add(Conv2D(32, (16,16), input_shape=(28, 28, 3), activation='relu', padding='same'))\nmodel.add(Conv2D(64, (8,8), input_shape=(28, 28, 3), activation='relu', padding='same'))\nmodel.add(Conv2D(64, (4,4), input_shape=(28, 28, 3), activation='relu', padding='same'))\nmodel.add(MaxPooling2D(pool_size=(4, 4)))\nmodel.add(Dropout(0.4))\nmodel.add(Flatten())\nmodel.add(Dense(128, activation='relu'))\nmodel.add(Dense(36, activation='softmax'))\n\nmodel.compile(loss='sparse_categorical_crossentropy', optimizer=optimizers.Adam(lr=0.0001), metrics=[custom_f1score])","metadata":{"id":"8IjCdBYrp4EK","execution":{"iopub.status.busy":"2023-11-25T06:16:45.786151Z","iopub.execute_input":"2023-11-25T06:16:45.786956Z","iopub.status.idle":"2023-11-25T06:16:45.897159Z","shell.execute_reply.started":"2023-11-25T06:16:45.786911Z","shell.execute_reply":"2023-11-25T06:16:45.896216Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"**Testing different Combinations**","metadata":{}},{"cell_type":"code","source":"K.clear_session()\nmodel = Sequential()\nmodel.add(Conv2D(16, (22,22), input_shape=(28, 28, 3), activation='relu', padding='same'))\nmodel.add(Conv2D(32, (16,16), activation='relu', padding='same'))\nmodel.add(Conv2D(64, (8,8), activation='relu', padding='same'))\nmodel.add(Conv2D(64, (4,4), activation='relu', padding='same'))\nmodel.add(MaxPooling2D(pool_size=(4, 4)))\nmodel.add(Dropout(0.4))\nmodel.add(Flatten())\nmodel.add(Dense(128, activation='relu'))\nmodel.add(Dense(36, activation='softmax'))\n\nmodel.compile(loss='sparse_categorical_crossentropy', optimizer=optimizers.Adam(lr=0.0001), metrics=[custom_f1score])","metadata":{"execution":{"iopub.status.busy":"2023-11-25T06:51:36.942356Z","iopub.execute_input":"2023-11-25T06:51:36.942855Z","iopub.status.idle":"2023-11-25T06:51:37.057767Z","shell.execute_reply.started":"2023-11-25T06:51:36.942813Z","shell.execute_reply":"2023-11-25T06:51:37.056580Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"model.summary()","metadata":{"execution":{"iopub.status.busy":"2023-11-25T06:16:51.246243Z","iopub.execute_input":"2023-11-25T06:16:51.246850Z","iopub.status.idle":"2023-11-25T06:16:51.258711Z","shell.execute_reply.started":"2023-11-25T06:16:51.246811Z","shell.execute_reply":"2023-11-25T06:16:51.257541Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"class stop_training_callback(tf.keras.callbacks.Callback):\n  def on_epoch_end(self, epoch, logs={}):\n    if(logs.get('val_custom_f1score') > 0.99):\n      self.model.stop_training = True","metadata":{"id":"w5aaqsHABUwx","execution":{"iopub.status.busy":"2023-11-25T06:16:53.986172Z","iopub.execute_input":"2023-11-25T06:16:53.986615Z","iopub.status.idle":"2023-11-25T06:16:53.994342Z","shell.execute_reply.started":"2023-11-25T06:16:53.986574Z","shell.execute_reply":"2023-11-25T06:16:53.993016Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"batch_size = 1\ncallbacks = [stop_training_callback()]\nmodel.fit_generator(\n      train_generator,\n      steps_per_epoch = train_generator.samples // batch_size,\n      validation_data = validation_generator, \n      epochs = 7, verbose=1, callbacks=callbacks)","metadata":{"id":"KPAtDd_Jp4BP","outputId":"77203b1e-c29d-408a-8ae4-8707569449a8","execution":{"iopub.status.busy":"2023-11-25T06:51:54.269882Z","iopub.execute_input":"2023-11-25T06:51:54.270507Z","iopub.status.idle":"2023-11-25T06:56:16.491585Z","shell.execute_reply.started":"2023-11-25T06:51:54.270422Z","shell.execute_reply":"2023-11-25T06:56:16.490557Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"model.save(\"model.h5\")","metadata":{"execution":{"iopub.status.busy":"2023-11-25T06:27:08.265639Z","iopub.execute_input":"2023-11-25T06:27:08.266041Z","iopub.status.idle":"2023-11-25T06:27:08.327071Z","shell.execute_reply.started":"2023-11-25T06:27:08.266006Z","shell.execute_reply":"2023-11-25T06:27:08.325837Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"# Predicting the output\ndef fix_dimension_FirstHalf(img): \n  new_img = np.zeros((28,28,3))\n  for i in range(3):\n    new_img[:,:,i] = img\n  return new_img\n  \ndef show_results_FirstHalf():\n    dic = {}\n    characters = '0123456789ABCDEFGHIJKLMNOPQRSTUVWXYZ'\n    for i,c in enumerate(characters):\n        dic[i] = c\n\n    output = []\n    for i,ch in enumerate(char_FirstHalf): #iterating over the characters\n        img_ = cv2.resize(ch, (28,28), interpolation=cv2.INTER_AREA)\n        img = fix_dimension_FirstHalf(img_)\n        img = img.reshape(1,28,28,3) #preparing image for the model\n        y_ = model.predict_classes(img)[0] #predicting the class\n        character = dic[y_] #\n        output.append(character) #storing the result in a list\n        \n    plate_number = ''.join(output)\n    \n    return plate_number\n\n# print(show_results())","metadata":{"id":"3PICNwtZUDAD","outputId":"dadfd56e-ecd7-419e-938a-4275f18261d6","execution":{"iopub.status.busy":"2023-11-25T06:51:40.710241Z","iopub.execute_input":"2023-11-25T06:51:40.710698Z","iopub.status.idle":"2023-11-25T06:51:40.724721Z","shell.execute_reply.started":"2023-11-25T06:51:40.710661Z","shell.execute_reply":"2023-11-25T06:51:40.722914Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"# Predicting the output\ndef fix_dimension_SecondHalf(img): \n  new_img = np.zeros((28,28,3))\n  for i in range(3):\n    new_img[:,:,i] = img\n  return new_img\n  \ndef show_results_SecondHalf():\n    dic = {}\n    characters = '0123456789ABCDEFGHIJKLMNOPQRSTUVWXYZ'\n    for i,c in enumerate(characters):\n        dic[i] = c\n\n    output = []\n    for i,ch in enumerate(char_SecondHalf): #iterating over the characters\n        img_ = cv2.resize(ch, (28,28), interpolation=cv2.INTER_AREA)\n        img = fix_dimension_SecondHalf(img_)\n        img = img.reshape(1,28,28,3)\n        y_ = model.predict_classes(img)[0]\n        character = dic[y_]\n        output.append(character)\n        \n    plate_number = ''.join(output)\n    \n    return plate_number\n\n# print(show_results())","metadata":{"execution":{"iopub.status.busy":"2023-11-25T06:51:44.510465Z","iopub.execute_input":"2023-11-25T06:51:44.510894Z","iopub.status.idle":"2023-11-25T06:51:44.525237Z","shell.execute_reply.started":"2023-11-25T06:51:44.510857Z","shell.execute_reply":"2023-11-25T06:51:44.524204Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"licensePlate = show_results_FirstHalf() + show_results_SecondHalf()\nlicensePlate","metadata":{"execution":{"iopub.status.busy":"2023-11-25T06:51:46.710295Z","iopub.execute_input":"2023-11-25T06:51:46.710748Z","iopub.status.idle":"2023-11-25T06:51:47.142187Z","shell.execute_reply.started":"2023-11-25T06:51:46.710708Z","shell.execute_reply":"2023-11-25T06:51:47.141114Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"file_path = '/kaggle/working/data.txt'  # Specify the file path\n\n# Open the file in write mode\nwith open(file_path, 'w') as file:\n    file.write(\"John Doe,ABC1293\\n\")\n    file.write(\"Jane Smith,LEB5700\\n\")\n    file.write(\"Mark Johnson,PQR4756\\n\")\n    file.write(\"Mohsin Ali,LEC4800\\n\")\n    file.write(\"Ahmad Jalal,XYZ7890\\n\")\n    file.write(\"Jalal Ali,PQR4563\\n\")\n    file.write(\"Ali Hassan,LEA4861\\n\")\n    file.write(\"Jameel,LER1234\\n\")","metadata":{"execution":{"iopub.status.busy":"2023-11-25T06:27:41.326387Z","iopub.execute_input":"2023-11-25T06:27:41.326854Z","iopub.status.idle":"2023-11-25T06:27:41.334917Z","shell.execute_reply.started":"2023-11-25T06:27:41.326815Z","shell.execute_reply":"2023-11-25T06:27:41.333779Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"def search_person_by_license_plate(file_path, license_plate):\n    with open(file_path, 'r') as file:\n        for line in file:\n            data = line.strip().split(',')\n            if data[1] == license_plate:\n                return data[0]  # Return the person's name\n    return None  # License plate not found\n\n# Specify the file path and license plate to search\nfile_path = 'data.txt'\nlicense_plate = licensePlate\n\n# Search for the person\nperson_name = search_person_by_license_plate(file_path, license_plate)\nif person_name:\n    print(f\"Person found: {person_name}\")\nelse:\n    print(\"Person not found\")","metadata":{"execution":{"iopub.status.busy":"2023-11-25T06:48:16.351082Z","iopub.execute_input":"2023-11-25T06:48:16.351571Z","iopub.status.idle":"2023-11-25T06:48:16.363290Z","shell.execute_reply.started":"2023-11-25T06:48:16.351529Z","shell.execute_reply":"2023-11-25T06:48:16.361588Z"},"trusted":true},"execution_count":null,"outputs":[]}]}