image detect stuff

backend/main.py

@@ -0,0 +1,95 @@
+# from flask import Flask
+
+# app = Flask(__name__)
+
+# @app.route("/")
+# def hello_world():
+#     return "<p>asdasd!</p>"
+
+
+#: flask to get data from frontend (assume we have data for now) (last thing we build in backend)
+#: OCR or Pytesseract for extracting text from image
+#: Pillow library to edit on the images
+#: googletrans to translate the text from language 1 to English
+#: 
+
+import cv2
+import pytesseract
+from googletrans import Translator
+from manga_ocr import MangaOcr
+from PIL import Image
+
+MOCKIMAGEPATH = 'japanese.png'
+mockImage = cv2.imread(MOCKIMAGEPATH)
+pytesseract_config = r'--oem 3 --psm 5 -l jpn_vert'
+
+grayscale = cv2.cvtColor(mockImage, cv2.COLOR_BGR2GRAY) # convert image to grayscale
+blur = cv2.GaussianBlur(grayscale, (9, 9), 0)
+# threshold = cv2.threshold(blur, 0, 255, cv2.THRESH_BINARY_INV + cv2.THRESH_OTSU)[1]
+
+# Attempt to remove noise from image
+# kernel = cv2.getStructuringElement(cv2.MORPH_RECT, (3, 3))
+# opening = cv2.morphologyEx(threshold, cv2.MORPH_OPEN, kernel, iterations=1)
+# invert = 255 - opening
+img = cv2.imread(MOCKIMAGEPATH)
+gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
+
+# Performing OTSU threshold
+ret, thresh1 = cv2.threshold(gray, 0, 255, cv2.THRESH_OTSU | cv2.THRESH_BINARY_INV)
+
+# Specify structure shape and kernel size.
+# Kernel size increases or decreases the area
+# of the rectangle to be detected.
+# A smaller value like (10, 10) will detect
+# each word instead of a sentence.
+rect_kernel = cv2.getStructuringElement(cv2.MORPH_RECT, (18, 18))
+
+# Applying dilation on the threshold image
+dilation = cv2.dilate(thresh1, rect_kernel, iterations = 1)
+cv2.imshow('dilation ', dilation)
+
+# Finding contours
+contours, hierarchy = cv2.findContours(dilation, cv2.RETR_EXTERNAL,
+                                                 cv2.CHAIN_APPROX_NONE)
+
+# Creating a copy of image
+im2 = img.copy()
+
+# A text file is created and flushed
+
+# Looping through the identified contours
+# Then rectangular part is cropped and passed on
+# to pytesseract for extracting text from it
+# Extracted text is then written into the text file
+for cnt in contours:
+    x, y, w, h = cv2.boundingRect(cnt)
+
+    # Drawing a rectangle on copied image
+    rect = cv2.rectangle(im2, (x, y), (x + w, y + h), (0, 255, 0), 2)
+    cv2.imshow('rectangle is ', rect)
+
+    # Cropping the text block for giving input to OCR
+    cropped = im2[y:y + h, x:x + w]
+
+    # Open the file in append mode
+    file = open("recognized.txt", "a")
+
+    # Apply OCR on the cropped image
+    text = pytesseract.image_to_string(cropped)
+    print(text)
+
+cv2.waitKey()
+# jp_text = pytesseract.image_to_string(invert, config=pytesseract_config)
+# print(jp_text)
+
+# cv2.imshow('mockimage', mockImage)
+# cv2.imshow('blur', blur)
+# cv2.imshow('threshold', threshold)
+# cv2.imshow('opening', opening)
+# cv2.imshow('invert', invert)
+# cv2.waitKey()
+
+# translator = Translator()
+# translatedText = translator.translate(jp_text, dest="en", src="ja")
+
+# print('[translated]: ', translatedText.text)

backend/requirements.txt

@@ -1 +1,7 @@
-flake8
+flake8
+Flask
+pytesseract
+numpy
+googletrans==4.0.0-rc1
+opencv-python
+manga-ocr

backend/test.txt

@@ -0,0 +1,78 @@
+== DEFAULT ==
+
+陣夫団…
+
+3J心惨枯コピ
+
+的S刀は8條パ避
+pn)っっべ直玩
+所0しWつろう@…
+
+== PSM 5 == 
+it           ョ
+アク   駐 Je転つじ      垂   
+4 。 六は衝
+5 病生4にだだなくり月還=     
+世-   的0し民つっ6G…   ミ  ミ
+ラー
+
+
+
+== JP text == 
+弄人団…
+
+32J心羽村つレ        
+
+的刀は6條で避        
+
+Jpいひいう)っうべ相対
+
+拘でしつう@…
+2
+
+== JP text == 
+レ
+則失団…
+2JD尺本つレ
+的や刀け6條で
+Jpいう)っつうっうべで直人
+拘でしつう@…
+
+
+
+冊挫団…
+2J①尺村つビレ     
+SS怠け6條%い      
+Jpいつつべ直志    
+拘でし8H6つろう@… 
+
+9
+
+Gaussian kernel = 5x5
+
+陣老団…
+2J①尺寺つビレ      
+的や台中6條る      
+jpいう)うつうべ牛装
+的しWOつっ@G…      
+
+に
+
+Gaussian kernel = 7x7
+
+陣挫団…
+2JQ尺相つレ
+Sや刀け16條%
+おpいつっうべ直人      
+的でしレH6つうG…       
+
+向
+
+Gaussian kernel = 9x9
+
+冊挫団…
+32J心尺相つレ
+的や刀はで條へ
+Jpいうつつうべ直乱      
+的でしレH6つっう@…      
+仁