Optical Character Recognition (OCR) is a technology used to extract text from images which is used in applications like document digitization, license plate recognition and automated data entry. In this article, we explore how to detect and extract text from images usingOpenCVfor image processing andTesseract OCRfor text recognition.

Before we start we need to install required libraries using following commands:

• !pip install opencv-python
• !pip install pytesseract
• !sudo apt-get install tesseract-ocr

Step 1: Importing Required Packages

Import the required Python libraries likeOpenCV,pytesseractandmatplotlib.

importcv2importpytesseractfrommatplotlibimportpyplotasplt

Step 2: Loading Image

Now we will load the image usingcv2.imread()function. OpenCV loads images in BGR format but matplotlib expects RGB format so we convert it usingcv2.cvtColor().

image_path="/mnt/data/08d702e9-fb88-4a5b-98ce-9ad0bc19afd9.png"image=cv2.imread(image_path)image_rgb=cv2.cvtColor(image,cv2.COLOR_BGR2RGB)

Step 3: Converting Image to Grayscale

Convert the input image to grayscale to simplify the image and remove color information.

gray=cv2.cvtColor(img,cv2.COLOR_BGR2GRAY)print("Grayscale Image:")cv2_imshow(gray)

Step 4:Displaying Original Image

Let's see original image with the help of matplotlib.

plt.figure(figsize=(10,6))plt.imshow(image_rgb)plt.title("Original Image")plt.axis("off")plt.show()

Output:

Step 5: Extracting Text from Image

In this step we extract text from the image with the help ofpytesseract.image_to_string().

extracted_text=pytesseract.image_to_string(image_rgb)print(" Extracted Text:\n")print(extracted_text)

Output:

Step 6: Drawing Bounding Boxes Around Detected Text

Now we loop through each detected word and draw ablue box around it usingcv2.reactangle.

data=pytesseract.image_to_data(image_rgb,output_type=pytesseract.Output.DICT)n_boxes=len(data['level'])foriinrange(n_boxes):(x,y,w,h)=(data['left'][i],data['top'][i],data['width'][i],data['height'][i])cv2.rectangle(image_rgb,(x,y),(x+w,y+h),(255,0,0),2)

Step 7: Showing Image with Bounding Boxes

Finally this shows the processed image with bounding boxes around the detected text.

plt.figure(figsize=(10,6))plt.imshow(image_rgb)plt.title("Image with Text Bounding Boxes")plt.axis("off")plt.show()

Output:

Get complete notebook link here :

Notebook:click here.
Sample Image:click here.

Similar Reads