Duplicate from ai-based/azure_ocr

Co-authored-by: Meet Jani <[email protected]>

.gitattributes

@@ -0,0 +1,34 @@
+*.7z filter=lfs diff=lfs merge=lfs -text
+*.arrow filter=lfs diff=lfs merge=lfs -text
+*.bin filter=lfs diff=lfs merge=lfs -text
+*.bz2 filter=lfs diff=lfs merge=lfs -text
+*.ckpt filter=lfs diff=lfs merge=lfs -text
+*.ftz filter=lfs diff=lfs merge=lfs -text
+*.gz filter=lfs diff=lfs merge=lfs -text
+*.h5 filter=lfs diff=lfs merge=lfs -text
+*.joblib filter=lfs diff=lfs merge=lfs -text
+*.lfs.* filter=lfs diff=lfs merge=lfs -text
+*.mlmodel filter=lfs diff=lfs merge=lfs -text
+*.model filter=lfs diff=lfs merge=lfs -text
+*.msgpack filter=lfs diff=lfs merge=lfs -text
+*.npy filter=lfs diff=lfs merge=lfs -text
+*.npz filter=lfs diff=lfs merge=lfs -text
+*.onnx filter=lfs diff=lfs merge=lfs -text
+*.ot filter=lfs diff=lfs merge=lfs -text
+*.parquet filter=lfs diff=lfs merge=lfs -text
+*.pb filter=lfs diff=lfs merge=lfs -text
+*.pickle filter=lfs diff=lfs merge=lfs -text
+*.pkl filter=lfs diff=lfs merge=lfs -text
+*.pt filter=lfs diff=lfs merge=lfs -text
+*.pth filter=lfs diff=lfs merge=lfs -text
+*.rar filter=lfs diff=lfs merge=lfs -text
+*.safetensors filter=lfs diff=lfs merge=lfs -text
+saved_model/**/* filter=lfs diff=lfs merge=lfs -text
+*.tar.* filter=lfs diff=lfs merge=lfs -text
+*.tflite filter=lfs diff=lfs merge=lfs -text
+*.tgz filter=lfs diff=lfs merge=lfs -text
+*.wasm filter=lfs diff=lfs merge=lfs -text
+*.xz filter=lfs diff=lfs merge=lfs -text
+*.zip filter=lfs diff=lfs merge=lfs -text
+*.zst filter=lfs diff=lfs merge=lfs -text
+*tfevents* filter=lfs diff=lfs merge=lfs -text

README.md

@@ -0,0 +1,14 @@
+---
+title: Azure Ocr
+emoji: 🏢
+colorFrom: yellow
+colorTo: green
+sdk: gradio
+sdk_version: 3.24.1
+app_file: app.py
+pinned: false
+license: other
+duplicated_from: ai-based/azure_ocr
+---
+
+Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference

app.py

@@ -0,0 +1,67 @@
+import os
+import openai
+import gradio as gr
+from azure.cognitiveservices.vision.computervision import ComputerVisionClient
+from msrest.authentication import CognitiveServicesCredentials
+from azure.storage.blob import BlobClient
+#import utils functions
+from preprocessing_images import preprocessing_function
+from extract_text import azure_ocr
+my_container = os.getenv("AZURE_CONTAINER")
+subscription_key = os.getenv("SUB_KEY")
+endpoint = os.getenv("AZURE_ENDPOINT")
+connection_string = os.getenv("AZURE_CON_STRING")
+openai.api_key = os.getenv("OPENAI_API_KEY")
+computervision_client = ComputerVisionClient(endpoint, CognitiveServicesCredentials(subscription_key))
+
+def ocr_pdf(pdf_url):
+    preprocessing_function(pdf_url)
+    my_blob = pdf_url.split('/')[-1]
+    blob = BlobClient.from_connection_string(conn_str=connection_string, container_name= my_container, blob_name=my_blob)
+    with open("answer_paper.pdf", "rb") as data:
+        blob.upload_blob(data,overwrite=True) 
+    text = azure_ocr(blob.url,computervision_client)
+    return text.strip()
+
+def classify_cause(incident_description):
+    response = openai.Completion.create(
+    engine="text-davinci-003",
+    prompt= f"Identify the root cause from the below list:\nincident_description:{incident_description}\n",
+    temperature= 0,
+    max_tokens= 50,
+    n=1,
+    stop=None
+    #timeout=15,
+    )
+    classification = response.choices[0].text.strip()
+    return classification
+    
+def classify_class(incident_description):
+    response = openai.Completion.create(
+    engine="text-davinci-003",
+    prompt= f"Classify the following incident description into one of the given classes:Aircraft Autopilot Problem, Auxiliary Power Problem,Cabin Pressure Problem, Engine Problem,Fuel System Problem,Avionics Problem,Communications Problem,Electrical System Problem,Engine Problem,Fire/Smoke Problem,Fuel System Problem,Ground Service Problem,Hydraulic System Problem,Ice/Frost Problem,Landing Gear Problem,Maintenance Problem,Oxygen System Problem,other problem\nincident_description:{incident_description}\n", 
+    temperature= 0,
+    max_tokens= 50,
+    n=1,
+    stop=None
+    #timeout=15,
+    )
+    classification = response.choices[0].text.strip()
+    return classification
+
+    
+def avatiation(pdf_url):
+    pdftext = ocr_pdf(pdf_url)
+    defect_class = classify_class(pdftext)
+    main_issue =  classify_cause(pdftext)
+    return defect_class, main_issue
+
+
+inputs = gr.inputs.Textbox(label="Link for aviation log reports")
+
+outputs = [gr.outputs.Textbox(label="Main Issue of the log report"),
+           gr.outputs.Textbox(label="category of the log report")]
+
+demo = gr.Interface(fn=avatiation,inputs=inputs,outputs=outputs, title="ATA Auto classification using OCR and GPT3 ")
+demo.launch()
+

extract_text.py

@@ -0,0 +1,47 @@
+from preprocessing_images import preprocessing_function
+from datetime import datetime
+from azure.storage.blob import BlobClient
+from msrest.authentication import CognitiveServicesCredentials
+#importing azure packages
+from azure.cognitiveservices.vision.computervision import ComputerVisionClient
+from azure.cognitiveservices.vision.computervision.models import OperationStatusCodes
+
+#ocr extraction using azure computer vision API
+def azure_ocr(pdf_url,computervision_client):
+    try:
+        read_response = computervision_client.read(pdf_url,raw=True)
+        read_operation_location = read_response.headers["Operation-Location"]
+        operation_id = read_operation_location.split("/")[-1]
+        while True:
+            read_result = computervision_client.get_read_result(operation_id)
+            if read_result.status not in ['notStarted', 'running']:
+                break
+        words = []
+        if read_result.status == OperationStatusCodes.succeeded:
+            for text_result in read_result.analyze_result.read_results:
+                for line in text_result.lines:
+                    words.append(line.text)           
+        all_text = ' '.join(words)  
+        return all_text
+    except Exception as e:
+        raise Exception(e)
+def extract_text_from_url(test_pdf_url):
+    try:
+        preprocessing_function(test_pdf_url)
+        my_blob = 'test_clean_pdf' + datetime.now().strftime('%Y_%m_%d_%H_%M_%S')
+        blob = BlobClient.from_connection_string(conn_str=connection_string, container_name= my_container, blob_name=my_blob)
+        with open("answer_paper.pdf", "rb") as data:
+            blob.upload_blob(data) 
+        computervision_client = ComputerVisionClient(endpoint, CognitiveServicesCredentials(subscription_key))
+        text = azure_ocr(blob.url, computervision_client)
+        text = text.lower()
+        n = text.find("150 word")
+        if n > 0:
+            text = text[n+10:]
+
+        elif text.find("150 ward") > 0:
+            nn = text.find("150 ward")
+            text = text[nn+10:]
+        return text
+    except Exception as e:
+        raise Exception(e)

packages.txt

@@ -0,0 +1 @@
+poppler-utils

preprocessing_images.py

@@ -0,0 +1,91 @@
+#importing packages
+from pdf2image import convert_from_path
+from fpdf import FPDF
+import cv2
+import numpy as np 
+import requests
+from PIL import Image
+from PIL import ImageEnhance
+
+def boxcutter(img):
+    img = np.array(img)
+    height, width = img.shape[:2]
+    gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY) # convert to grayscale
+    blurred = cv2.GaussianBlur(gray, (5, 5), 0)
+    (T, threshInv) = cv2.threshold(blurred, 230, 255, cv2.THRESH_BINARY_INV)
+    contours, hierarchy = cv2.findContours(threshInv,cv2.RETR_LIST, \
+                                       cv2.CHAIN_APPROX_SIMPLE)
+    mx = (0,0,0,0)      # biggest bounding box so far
+    mx_area =0
+    my_dict = {}
+    for cont in contours:
+        x,y,w,h = cv2.boundingRect(cont)
+        area = w*h
+        nw = x,y,w,h
+        my_dict[nw] = area
+    if len(my_dict)<=2:
+        return img
+    else:
+        x,y,w,h = sorted(my_dict, key=my_dict.get)[-2]
+        roi=img[y:y+h,x:x+w]
+        height_r, width_r = roi.shape[:2]
+        if height_r > int(height*0.6) and width_r > int(width*0.6):
+            return roi
+    #         cv2.imwrite('C:/Users/Meet/Desktop/crop/'+ img_path.split('/')[-1], roi)
+    #         cv2.rectangle(img,(x,y),(x+w,y+h),(200,0,0),2)
+    #         cv2.imwrite('C:/Users/Meet/Desktop/cont/' + img_path.split('/')[-1], img)
+
+        else:
+            return img
+    #         cv2.imwrite('C:/Users/Meet/Desktop/crop/'+ img_path.split('/')[-1], img)
+    #         cv2.rectangle(img,(x,y),(x+w,y+h),(200,0,0),2)
+    #         cv2.imwrite('C:/Users/Meet/Desktop/cont/' + img_path.split('/')[-1], img)
+
+def noise_removal(image):
+    #increasing contrast of text
+    image=Image.fromarray(image)
+    enhancer=ImageEnhance.Contrast(image)
+    enhanced_image=enhancer.enhance(2.5)
+    image=np.array(enhanced_image)
+    #noise removel using dilate,erode, morphology and median blur
+    kernel = np.ones((1, 1), np.uint8)
+    image=cv2.dilate(image,kernel,iterations=1)
+    kernel=np.ones((1,1),np.uint8)
+    image=cv2.erode(image,kernel,iterations=1)
+    image = cv2.morphologyEx(image, cv2.MORPH_CLOSE, kernel,iterations=1)
+    image=cv2.medianBlur(image,3)
+    image=cv2.bitwise_not(image)
+    kernel=np.ones((1,1),np.uint8)
+    image=cv2.erode(image,kernel,iterations=3)
+    image=cv2.bitwise_not(image)
+    image=cv2.bitwise_not(image)
+    kernel=np.ones((1,1),np.uint8)
+    image=cv2.dilate(image,kernel,iterations=3)
+    image=cv2.bitwise_not(image)
+    return (image)
+
+def preprocessing_function(url):
+    #reading and converting the pdf into image format
+    response = requests.get(url)
+    with open('metadata.pdf', 'wb') as f:
+        f.write(response.content)
+    images = convert_from_path('metadata.pdf', 200)
+    #removing red color from the answer paper
+    for i in range(len(images)):
+        img = boxcutter(images[i])
+        image = cv2.cvtColor(img, cv2.COLOR_RGB2BGR)
+        img_hsv=cv2.cvtColor(image, cv2.COLOR_BGR2HSV)
+        lower_red = np.array([161,50,50])
+        upper_red = np.array([189,255,255])
+        mask1 = cv2.inRange(img_hsv, lower_red, upper_red)
+        image[np.where(mask1==255)] = 255
+        #using noise removal function to remove noise and enhance the images
+        img_bw = cv2.cvtColor(image,cv2.COLOR_BGR2GRAY)
+        no_noise = noise_removal(img_bw)
+        cv2.imwrite("no_noise"+str(i)+".jpg", no_noise)
+    #saving the the final preprocessed images as pdf    
+    pdf = FPDF()
+    for j in range(len(images)):
+        pdf.add_page()
+        pdf.image("no_noise"+str(j)+".jpg", 0, 0, 200, 300)
+    pdf.output("answer_paper.pdf", "F")

requirements.txt

@@ -0,0 +1,25 @@
+absl-py==1.0.0
+astunparse==1.6.3
+azure-cognitiveservices-vision-computervision==0.9.0
+azure-common==1.1.27
+azure-core==1.21.1
+azure-storage-blob==12.9.0
+flatbuffers==2.0
+fpdf==1.7.2
+h5py==3.6.0
+jsonify==0.5
+joblib==1.1.0
+numpy==1.22.1
+msrest==0.6.21
+nltk==3.7
+opencv-python-headless==4.6.0.66
+Pillow==9.2.0
+pdf2image==1.16.0
+PyPDF2==2.9.0
+pandas==1.4.3
+requests==2.28.1
+scikit-learn==1.1.1
+scipy==1.8.1
+urllib3==1.26.11
+gradio==3.26.0
+openai==0.27.8