Modified code and README

KananVyas · Jun 29, 2020 · 853c973 · 853c973
1 parent 4d0aa27
commit 853c973
Show file tree

Hide file tree

Showing 17 changed files with 36 additions and 18 deletions.
diff --git a/Image_bin.jpg → Images/Image_bin.jpg b/Image_bin.jpg → Images/Image_bin.jpg
diff --git a/horizontal_lines.jpg → Images/horizontal_lines.jpg b/horizontal_lines.jpg → Images/horizontal_lines.jpg
diff --git a/img_final_bin.jpg → Images/img_final_bin.jpg b/img_final_bin.jpg → Images/img_final_bin.jpg
diff --git a/verticle_lines.jpg → Images/verticle_lines.jpg b/verticle_lines.jpg → Images/verticle_lines.jpg
diff --git a/Cropped/1.png → Output/1.png b/Cropped/1.png → Output/1.png
diff --git a/Cropped/10.png → Output/10.png b/Cropped/10.png → Output/10.png
diff --git a/Cropped/2.png → Output/2.png b/Cropped/2.png → Output/2.png
diff --git a/Cropped/3.png → Output/3.png b/Cropped/3.png → Output/3.png
diff --git a/Cropped/4.png → Output/4.png b/Cropped/4.png → Output/4.png
diff --git a/Cropped/5.png → Output/5.png b/Cropped/5.png → Output/5.png
diff --git a/Cropped/6.png → Output/6.png b/Cropped/6.png → Output/6.png
diff --git a/Cropped/7.png → Output/7.png b/Cropped/7.png → Output/7.png
diff --git a/Cropped/8.png → Output/8.png b/Cropped/8.png → Output/8.png
diff --git a/Cropped/9.png → Output/9.png b/Cropped/9.png → Output/9.png
diff --git a/README.md b/README.md
@@ -1,14 +1,22 @@
+
+# A Box detection algorithm for tabularized data.
 
-When you are working with Optical character recognition(OCR) or any data or 
-object recognition problem, the first thing to do is preprocessing. Here preprocessing means
-to extract the location where our information is located. After extracting the location,
-any machine algorithm will be performed on that image. 
+ - This code is used to extract data which is in tabular format using image processing techniques.
+ - When you are working with Optical character recognition(OCR) or any data or object recognition problem, the first thing to do is preprocessing. 
+ - Here preprocessing means to extract the box where our data is located. After extracting the boxes, any OCR algorithm can be performed on those crops for recognition.
+ ![alt text](http://github.com/KananVyas/BoxDetection/blob/master/41.jpg)
 
-The problem arises when you have to detect objects which are located in any tables/boxes or
-in row-column format. If the image is like this then you have to detect boxes and extract them one by one. 
+ - The problem arises when you have to detect objects which are located  in any tables/boxes or in row-column format. If the image is like this then you have to detect boxes and extract them one by one.
 Now it should be done accurately for all images.
 
-This algorithm helps to detect every boxes accurately and save it in a "Cropped" folder.The code is shown in box_detection.py 
-and the test image is "41.jpg".
+
 
-you can see the medium blog for this code: https://medium.com/@kananvyas/a-box-detection-algorithm-for-any-image-containing-boxes-756c15d7ed26
+This algorithm helps to detect every boxes accurately and save it in a `/Output/` folder. The code is shown in `src/box_detection.py`
+
+**USAGE:**
+ - Run `python src/box_detection.py`
+ - You can see the output crops on `/Output` folder
+ ![alt text](http://github.com/KananVyas/BoxDetection/blob/master/Output/1.png)
+ ![alt text](http://github.com/KananVyas/BoxDetection/blob/master/Output/2.png)
+
+You can also read the medium article for understanding about the algorithm: https://medium.com/@kananvyas/a-box-detection-algorithm-for-any-image-containing-boxes-756c15d7ed26
diff --git a/box_detection.ipynb → src/box_detection.ipynb b/box_detection.ipynb → src/box_detection.ipynb
@@ -48,7 +48,7 @@
     "(thresh, img_bin) = cv2.threshold(img, 128, 255,cv2.THRESH_BINARY|     cv2.THRESH_OTSU)\n",
     "# Invert the image\n",
     "img_bin = 255-img_bin \n",
-    "cv2.imwrite(\"Image_bin.jpg\",img_bin)\n",
+    "cv2.imwrite(\"Images/Image_bin.jpg\",img_bin)\n",
     "plt.axis('off')\n",
     "plt.imshow(img_bin);"
    ]
@@ -88,7 +88,7 @@
     "    # Morphological operation to detect vertical lines from an image\n",
     "    img_temp1 = cv2.erode(img_bin, verticle_kernel, iterations=3)\n",
     "    verticle_lines_img = cv2.dilate(img_temp1, verticle_kernel, iterations=3)\n",
-    "    cv2.imwrite(\"verticle_lines.jpg\",verticle_lines_img)\n",
+    "    cv2.imwrite(\"Images/verticle_lines.jpg\",verticle_lines_img)\n",
     "    # Morphological operation to detect horizontal lines from an image\n",
     "    img_temp2 = cv2.erode(img_bin, hori_kernel, iterations=3)\n",
     "    horizontal_lines_img = cv2.dilate(img_temp2, hori_kernel, iterations=3)\n",
@@ -240,4 +240,4 @@
  },
  "nbformat": 4,
  "nbformat_minor": 2
-}
+}
diff --git a/box_detection.py → src/box_detection.py b/box_detection.py → src/box_detection.py
@@ -24,15 +24,20 @@ def sort_contours(cnts, method="left-to-right"):
     # return the list of sorted contours and bounding boxes
     return (cnts, boundingBoxes)
 
+
+#Functon for extracting the box
 def box_extraction(img_for_box_extraction_path, cropped_dir_path):
 
+    print("Reading image..")
     img = cv2.imread(img_for_box_extraction_path, 0)  # Read the image
     (thresh, img_bin) = cv2.threshold(img, 128, 255,
                                       cv2.THRESH_BINARY | cv2.THRESH_OTSU)  # Thresholding the image
     img_bin = 255-img_bin  # Invert the image
 
-    cv2.imwrite("Image_bin.jpg",img_bin)
-
+    print("Storing binary image to Images/Image_bin.jpg..")
+    cv2.imwrite("Images/Image_bin.jpg",img_bin)
+
+    print("Applying Morphological Operations..")
     # Defining a kernel length
     kernel_length = np.array(img).shape[1]//40
 
@@ -46,12 +51,12 @@ def box_extraction(img_for_box_extraction_path, cropped_dir_path):
     # Morphological operation to detect verticle lines from an image
     img_temp1 = cv2.erode(img_bin, verticle_kernel, iterations=3)
     verticle_lines_img = cv2.dilate(img_temp1, verticle_kernel, iterations=3)
-    cv2.imwrite("verticle_lines.jpg",verticle_lines_img)
+    cv2.imwrite("Images/verticle_lines.jpg",verticle_lines_img)
 
     # Morphological operation to detect horizontal lines from an image
     img_temp2 = cv2.erode(img_bin, hori_kernel, iterations=3)
     horizontal_lines_img = cv2.dilate(img_temp2, hori_kernel, iterations=3)
-    cv2.imwrite("horizontal_lines.jpg",horizontal_lines_img)
+    cv2.imwrite("Images/horizontal_lines.jpg",horizontal_lines_img)
 
     # Weighting parameters, this will decide the quantity of an image to be added to make a new image.
     alpha = 0.5
@@ -63,13 +68,16 @@ def box_extraction(img_for_box_extraction_path, cropped_dir_path):
 
     # For Debugging
     # Enable this line to see verticle and horizontal lines in the image which is used to find boxes
-    cv2.imwrite("img_final_bin.jpg",img_final_bin)
+    print("Binary image which only contains boxes: Images/img_final_bin.jpg")
+    cv2.imwrite("Images/img_final_bin.jpg",img_final_bin)
     # Find contours for image, which will detect all the boxes
     contours, hierarchy = cv2.findContours(
         img_final_bin, cv2.RETR_TREE, cv2.CHAIN_APPROX_SIMPLE)
     # Sort all the contours by top to bottom.
     (contours, boundingBoxes) = sort_contours(contours, method="top-to-bottom")
 
+    print("Output stored in Output directiory!")
+
     idx = 0
     for c in contours:
         # Returns the location and width,height for every contour
@@ -87,4 +95,6 @@ def box_extraction(img_for_box_extraction_path, cropped_dir_path):
     # cv2.imwrite("./Temp/img_contour.jpg", img)
 
 
-box_extraction("41.jpg", "./Cropped/")
+
+#Input image path and out folder
+box_extraction("41.jpg", "./Output/")