\n\n\n\n
Introduction to OCR<\/h2>\n\n\n\n
We have always seen Optical character recognition<\/strong> being used a lot in machine learning and deep learning. One of many such applications includes the identification and reading of credit cards and the card number.<\/p>\n\n\n\n The question that might come to your mind is Why? So this application could be of great help to banks and other financial institutions to digitally recognize the card numbers and type of card.<\/p>\n\n\n\n Now that we have understood the concept and what we are going to build by end of this tutorial.<\/p>\n\n\n\n Let’s start building the project step by step.<\/p>\n\n\n\n We’ll be working with numpy<\/a> and matplotlib<\/a> along with the openCV module in this case. <\/p>\n\n\n The card type is assigned according to the first digit of the card number. The same is displayed below.<\/p>\n\n\n In order to read the reference OCR image, we make use of the The pre-processing of the image includes converting it to a gray image and then thresholding + inverting the image to get the binary inverted image.<\/p>\n\n\n In this step, we find the contours present in the pre-processed image and then store the returned contour information. Next, we sort the contours from left-to-right as well as initialize a dictionary, digits, which map the digit name to the region of interest.<\/p>\n\n\n Now in this step, we loop through the image contours obtained in the previous step where each value holds the digit\/number along with the contour information. We further compute a bounding box for each contour and store the (x, y) coordinates along with the height and width of the box computed.<\/p>\n\n\n In this step, we load our photo of the credit card and then This step is followed by converting the image to The next step is to compute a Now this normalizing of values take place by computing the minimum and maximum value of the gradX and form an equation to achieve Finally, we find the Next, we\u2019ll sort the groupings<\/strong> from Now that we know where each group of four digits is, let\u2019s loop through the four sorted groupings and determine the digits therein. The looping includes thresholding, detecting contours, and template matching<\/strong> as well.<\/p>\n\n\n The code below will display the final Card Type, Card Number, and the OCR applied image.<\/p>\n\n\n Now let’s look at some sample outputs after implementing the code mentioned above on various credit card images.<\/p>\n\n\n\n I hope you understood the concept and loved the outputs. Try out the same with more images and get amazed with the results.<\/p>\n\n\n\n Happy Coding! 😇<\/p>\n\n\n\n Want to learn more? Check out the tutorials mentioned below:<\/p>\n\n\n\n The purpose of this tutorial is to help you build a credit card reader with OpenCV and machine learning techniques to identify the card number and the card type. Let us get started! Also read: How to read Images in Python using OpenCV? Introduction to OCR We have always seen Optical character recognition being used […]<\/p>\n","protected":false},"author":28,"featured_media":21938,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[9],"tags":[],"class_list":["post-21936","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-examples"],"blocksy_meta":[],"_links":{"self":[{"href":"https:\/\/www.askpython.com\/wp-json\/wp\/v2\/posts\/21936","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.askpython.com\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.askpython.com\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.askpython.com\/wp-json\/wp\/v2\/users\/28"}],"replies":[{"embeddable":true,"href":"https:\/\/www.askpython.com\/wp-json\/wp\/v2\/comments?post=21936"}],"version-history":[{"count":0,"href":"https:\/\/www.askpython.com\/wp-json\/wp\/v2\/posts\/21936\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.askpython.com\/wp-json\/wp\/v2\/media\/21938"}],"wp:attachment":[{"href":"https:\/\/www.askpython.com\/wp-json\/wp\/v2\/media?parent=21936"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.askpython.com\/wp-json\/wp\/v2\/categories?post=21936"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.askpython.com\/wp-json\/wp\/v2\/tags?post=21936"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}![\"Credit](\"https:\/\/www.askpython.com\/wp-content\/uploads\/2021\/09\/Credit_Card_Reader_Demonstration.jpg\")
\n\n\n\nImplementation of a Credit Card Reader in Python<\/h2>\n\n\n\n
\n\n\n\nStep 1: Importing Modules<\/h3>\n\n\n\n
\nimport cv2\nimport imutils\nimport argparse\nimport numpy as np\nfrom imutils import contours\nfrom matplotlib import pyplot as plt\n<\/pre><\/div>\n\n\n
\n\n\n\nStep 2: Assign Card type<\/h3>\n\n\n\n
\nFIRST_NUMBER = {\n "3": "American Express",\n "4": "Visa",\n "5": "MasterCard",\n "6": "Discover Card"}\n<\/pre><\/div>\n\n\n
\n\n\n\nStep 3: Loading and pre-processing of reference Image<\/h3>\n\n\n\n
imread<\/code><\/a> function. The reference image contains the digits 0-9 in the OCR-A font which can be later on used to perform matching later in the pipeline.<\/p>\n\n\n\n\nref = cv2.imread('ocr_a_reference.png')\nref = cv2.cvtColor(ref, cv2.COLOR_BGR2GRAY)\nref = cv2.threshold(ref, 10, 255, cv2.THRESH_BINARY_INV)[1]\n<\/pre><\/div>\n\n\n
\n\n\n\nStep 4: Detecting Contours<\/h3>\n\n\n\n
\nrefCnts = cv2.findContours(ref.copy(), cv2.RETR_EXTERNAL,cv2.CHAIN_APPROX_SIMPLE)\nrefCnts = imutils.grab_contours(refCnts)\nrefCnts = contours.sort_contours(refCnts, method="left-to-right")[0]\ndigits = {}\n<\/pre><\/div>\n\n\n
\n\n\n\nStep 5: Creating Bounding Boxes around Digits<\/h3>\n\n\n\n
\nfor (i, c) in enumerate(refCnts):\n (x, y, w, h) = cv2.boundingRect(c)\n roi = ref[y:y + h, x:x + w]\n roi = cv2.resize(roi, (57, 88))\n digits[i] = roi\n \nrectKernel = cv2.getStructuringElement(cv2.MORPH_RECT, (9, 3))\nsqKernel = cv2.getStructuringElement(cv2.MORPH_RECT, (5, 5))\n<\/pre><\/div>\n\n\n
\n\n\n\nStep 6: Loading and Pre-processing of the Credit card Image<\/h3>\n\n\n\n
resize<\/code> the same to the width of 300 in order to maintain the aspect ratio.<\/p>\n\n\n\ngrayscale<\/code>. After this, we perform morphological operations<\/code> on the grayscale image.<\/p>\n\n\n\nScharr gradient<\/code><\/strong> and store the result as gradX<\/code>. Then we calculate the absolute value of the gradX array stored. We aim to scale all the values in the range of 0-255<\/code><\/strong>.<\/p>\n\n\n\nmin-max normalization<\/code><\/strong>.<\/p>\n\n\n\ncontours<\/code> and store them in a list and initialize a list to hold the digit group locations. Then loop through the contours the same way we did for the reference image in step 5<\/code>.<\/p>\n\n\n\nleft to right<\/code> and initialize a list for the credit card digits.<\/p>\n\n\n\nimage = cv2.imread('credit_card_03.png')\nimage = imutils.resize(image, width=300)\ngray = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)\n\ntophat = cv2.morphologyEx(gray, cv2.MORPH_TOPHAT, rectKernel)\n\ngradX = cv2.Sobel(tophat, ddepth=cv2.CV_32F, dx=1, dy=0, ksize=-1)\ngradX = np.absolute(gradX)\n(minVal, maxVal) = (np.min(gradX), np.max(gradX))\ngradX = (255 * ((gradX - minVal) \/ (maxVal - minVal)))\ngradX = gradX.astype("uint8")\n\ngradX = cv2.morphologyEx(gradX, cv2.MORPH_CLOSE, rectKernel)\nthresh = cv2.threshold(gradX, 0, 255, cv2.THRESH_BINARY | cv2.THRESH_OTSU)[1]\n\nthresh = cv2.morphologyEx(thresh, cv2.MORPH_CLOSE, sqKernel)\n\ncnts = cv2.findContours(thresh.copy(), cv2.RETR_EXTERNAL,\n\tcv2.CHAIN_APPROX_SIMPLE)\ncnts = imutils.grab_contours(cnts)\nlocs = []\n\nfor (i, c) in enumerate(cnts):\n\t(x, y, w, h) = cv2.boundingRect(c)\n\tar = w \/ float(h)\n\tif ar > 2.5 and ar < 4.0:\n\t\tif (w > 40 and w < 55) and (h > 10 and h < 20):\n\t\t\tlocs.append((x, y, w, h))\n\nlocs = sorted(locs, key=lambda x:x[0])\noutput = []\n<\/pre><\/div>\n\n\n\nfor (i, (gX, gY, gW, gH)) in enumerate(locs):\n groupOutput = []\n group = gray[gY - 5:gY + gH + 5, gX - 5:gX + gW + 5]\n group = cv2.threshold(group, 0, 255, cv2.THRESH_BINARY | cv2.THRESH_OTSU)[1]\n digitCnts = cv2.findContours(group.copy(), cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_SIMPLE)\n digitCnts = imutils.grab_contours(digitCnts)\n digitCnts = contours.sort_contours(digitCnts, method="left-to-right")[0]\n for c in digitCnts:\n (x, y, w, h) = cv2.boundingRect(c)\n roi = group[y:y + h, x:x + w]\n roi = cv2.resize(roi, (57, 88))\n scores = []\n for (digit, digitROI) in digits.items():\n result = cv2.matchTemplate(roi, digitROI, cv2.TM_CCOEFF)\n (_, score, _, _) = cv2.minMaxLoc(result) \n scores.append(score)\n groupOutput.append(str(np.argmax(scores)))\n cv2.rectangle(image, (gX - 5, gY - 5),\n (gX + gW + 5, gY + gH + 5), (0, 0, 255), 2)\n cv2.putText(image, "".join(groupOutput), (gX, gY - 15),\n cv2.FONT_HERSHEY_SIMPLEX, 0.65, (0, 0, 255), 2)\n output.extend(groupOutput)\n<\/pre><\/div>\n\n\n
\n\n\n\nStep 7: Displaying the Final Results<\/h3>\n\n\n\n
\nprint("Credit Card Type: {}".format(FIRST_NUMBER[output[0]]))\nprint("Credit Card #: {}".format("".join(output)))\n\nplt.imshow(cv2.cvtColor(image, cv2.COLOR_BGR2RGB))\nplt.title('Image'); plt.show()\n<\/pre><\/div>\n\n\n
\n\n\n\nThe Final Code<\/h2>\n\n\n
\nimport cv2\nimport imutils\nimport argparse\nimport numpy as np\nfrom imutils import contours\nfrom matplotlib import pyplot as plt\n\nFIRST_NUMBER = {\n "3": "American Express",\n "4": "Visa",\n "5": "MasterCard",\n "6": "Discover Card"}\n\nref = cv2.imread('ocr_a_reference.png')\nref = cv2.cvtColor(ref, cv2.COLOR_BGR2GRAY)\nref = cv2.threshold(ref, 10, 255, cv2.THRESH_BINARY_INV)[1]\n\nrefCnts = cv2.findContours(ref.copy(), cv2.RETR_EXTERNAL,cv2.CHAIN_APPROX_SIMPLE)\nrefCnts = imutils.grab_contours(refCnts)\nrefCnts = contours.sort_contours(refCnts, method="left-to-right")[0]\ndigits = {}\n\nfor (i, c) in enumerate(refCnts):\n (x, y, w, h) = cv2.boundingRect(c)\n roi = ref[y:y + h, x:x + w]\n roi = cv2.resize(roi, (57, 88))\n digits[i] = roi\n \nrectKernel = cv2.getStructuringElement(cv2.MORPH_RECT, (9, 3))\nsqKernel = cv2.getStructuringElement(cv2.MORPH_RECT, (5, 5))\n\nimage = cv2.imread('credit_card_03.png')\nimage = imutils.resize(image, width=300)\ngray = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)\n\ntophat = cv2.morphologyEx(gray, cv2.MORPH_TOPHAT, rectKernel)\n\ngradX = cv2.Sobel(tophat, ddepth=cv2.CV_32F, dx=1, dy=0, ksize=-1)\ngradX = np.absolute(gradX)\n(minVal, maxVal) = (np.min(gradX), np.max(gradX))\ngradX = (255 * ((gradX - minVal) \/ (maxVal - minVal)))\ngradX = gradX.astype("uint8")\n\ngradX = cv2.morphologyEx(gradX, cv2.MORPH_CLOSE, rectKernel)\nthresh = cv2.threshold(gradX, 0, 255, cv2.THRESH_BINARY | cv2.THRESH_OTSU)[1]\n\nthresh = cv2.morphologyEx(thresh, cv2.MORPH_CLOSE, sqKernel)\n\ncnts = cv2.findContours(thresh.copy(), cv2.RETR_EXTERNAL,\n\tcv2.CHAIN_APPROX_SIMPLE)\ncnts = imutils.grab_contours(cnts)\nlocs = []\n\nfor (i, c) in enumerate(cnts):\n\t(x, y, w, h) = cv2.boundingRect(c)\n\tar = w \/ float(h)\n\tif ar > 2.5 and ar < 4.0:\n\t\tif (w > 40 and w < 55) and (h > 10 and h < 20):\n\t\t\tlocs.append((x, y, w, h))\n\nlocs = sorted(locs, key=lambda x:x[0])\noutput = []\n\nfor (i, (gX, gY, gW, gH)) in enumerate(locs):\n groupOutput = []\n group = gray[gY - 5:gY + gH + 5, gX - 5:gX + gW + 5]\n group = cv2.threshold(group, 0, 255, cv2.THRESH_BINARY | cv2.THRESH_OTSU)[1]\n digitCnts = cv2.findContours(group.copy(), cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_SIMPLE)\n digitCnts = imutils.grab_contours(digitCnts)\n digitCnts = contours.sort_contours(digitCnts, method="left-to-right")[0]\n for c in digitCnts:\n (x, y, w, h) = cv2.boundingRect(c)\n roi = group[y:y + h, x:x + w]\n roi = cv2.resize(roi, (57, 88))\n scores = []\n for (digit, digitROI) in digits.items():\n result = cv2.matchTemplate(roi, digitROI, cv2.TM_CCOEFF)\n (_, score, _, _) = cv2.minMaxLoc(result) \n scores.append(score)\n groupOutput.append(str(np.argmax(scores)))\n cv2.rectangle(image, (gX - 5, gY - 5),\n (gX + gW + 5, gY + gH + 5), (0, 0, 255), 2)\n cv2.putText(image, "".join(groupOutput), (gX, gY - 15),\n cv2.FONT_HERSHEY_SIMPLEX, 0.65, (0, 0, 255), 2)\n output.extend(groupOutput)\n \nprint("Credit Card Type: {}".format(FIRST_NUMBER[output[0]]))\nprint("Credit Card #: {}".format("".join(output)))\n\nplt.imshow(cv2.cvtColor(image, cv2.COLOR_BGR2RGB))\nplt.title('Image'); plt.show()\n<\/pre><\/div>\n\n\n
\n\n\n\nSome Sample Outputs<\/h2>\n\n\n\n
![\"Credit](\"https:\/\/www.askpython.com\/wp-content\/uploads\/2021\/09\/credit_card_Reader_Output_01.png\")
![\"Credit](\"https:\/\/www.askpython.com\/wp-content\/uploads\/2021\/09\/credit_card_Reader_Output_02.png\")
![\"Credit](\"https:\/\/www.askpython.com\/wp-content\/uploads\/2021\/09\/credit_card_Reader_Output_03.png\")
\n\n\n\nConclusion<\/h2>\n\n\n\n
\n","protected":false},"excerpt":{"rendered":"