From fe2c1025b102bbf20c9afbc29eedf7a5f069410c Mon Sep 17 00:00:00 2001
From: Samo Penic <samo.penic@gmail.com>
Date: Sat, 17 Nov 2018 14:11:17 +0000
Subject: [PATCH] Readme requirements.
---
Ocr.py | 95 +++++++++++++++++++++++++++++++++++++++++++----
1 files changed, 86 insertions(+), 9 deletions(-)
diff --git a/Ocr.py b/Ocr.py
index 0da4497..6e9a082 100644
--- a/Ocr.py
+++ b/Ocr.py
@@ -1,17 +1,20 @@
from pyzbar.pyzbar import decode
-from sid_process import enhanceSID
+from sid_process import getSID
import cv2
import numpy as np
import math
class Paper:
- def __init__(self, filename=None):
+ def __init__(self, filename=None, sid_classifier=None, settings=None):
self.filename = filename
self.invalid = None
self.QRData = None
+ self.settings = {"answer_threshold": 0.25} if settings is None else settings
self.errors = []
self.warnings = []
+ self.sid = None
+ self.sid_classifier = sid_classifier
if filename is not None:
self.loadImage(filename)
self.runOcr()
@@ -54,22 +57,30 @@
self.data = None
self.invalid = True
return
+ if(len(d)>1): #if there are multiple codes, get first ean or qr code available.
+ for dd in d:
+ if(dd.type=="EAN13" or dd.type=="QR"):
+ d[0]=dd
+ break
self.QRDecode = d
self.QRData = d[0].data
xpos = d[0].rect.left
ypos = d[0].rect.top
# check if image is rotated wrongly
- if xpos > self.imgHeight / 2.0 and ypost > self.imgWidth / 2.0:
+ if xpos > self.imgHeight / 2.0 and ypos > self.imgWidth / 2.0:
self.rotateAngle(180)
def rotateAngle(self, angle=0):
+ # rot_mat = cv2.getRotationMatrix2D(
+ # (self.imgHeight / 2, self.imgWidth / 2), angle, 1.0
+ # )
rot_mat = cv2.getRotationMatrix2D(
- (self.imgHeight / 2, self.imgWidth / 2), angle, 1.0
+ (self.imgWidth / 2, self.imgHeight / 2), angle, 1.0
)
result = cv2.warpAffine(
self.img,
rot_mat,
- (self.imgHeight, self.imgWidth),
+ (self.imgWidth, self.imgHeight),
flags=cv2.INTER_CUBIC,
borderMode=cv2.BORDER_CONSTANT,
borderValue=(255, 255, 255),
@@ -137,8 +148,8 @@
loc_filtered_x, loc_filtered_y = zip(
*sorted(zip(loc_filtered_x, loc_filtered_y))
)
- # loc=[loc_filtered_y,loc_filtered_x]
- # remove duplicates
+ # loc=[loc_filtered_y,loc_filtered_x]
+ # remove duplicates
a = np.diff(loc_filtered_x) > 40
a = np.append(a, True)
loc_filtered_x = np.array(loc_filtered_x)
@@ -213,5 +224,71 @@
self.answerMatrix.append(oneline)
def get_enhanced_sid(self):
- es= enhanceSID(self.img[int(0.04*self.imgHeight):int(0.08*self.imgHeight), int(0.7*self.imgWidth):int(0.99*self.imgWidth)])
- cv2.imwrite("enhancedSID.png",es)
\ No newline at end of file
+ if self.sid_classifier is None:
+ return "x"
+ if self.settings is not None:
+ sid_mask = self.settings.get("sid_mask", None)
+ es, err, warn = getSID(
+ self.img[
+ int(0.04 * self.imgHeight) : int(0.095 * self.imgHeight),
+ int(0.7 * self.imgWidth) : int(0.99 * self.imgWidth),
+ ],
+ self.sid_classifier,
+ sid_mask,
+ )
+ [self.errors.append(e) for e in err]
+ [self.warnings.append(w) for w in warn]
+ return es
+
+ def get_code_data(self):
+ if self.QRData is None:
+ self.errors.append("Could not read QR or EAN code! Not an exam?")
+ retval = {
+ "exam_id": None,
+ "page_no": None,
+ "paper_id": None,
+ "faculty_id": None,
+ "sid": None,
+ }
+ return retval
+ qrdata = bytes.decode(self.QRData, "utf8")
+ if self.QRDecode[0].type == "EAN13":
+ return {
+ "exam_id": int(qrdata[0:7]),
+ "page_no": int(qrdata[7]),
+ "paper_id": int(qrdata[-5:-1]),
+ "faculty_id": None,
+ "sid": None,
+ }
+ else:
+ data = qrdata.split(",")
+ retval = {
+ "exam_id": int(data[1]),
+ "page_no": int(data[3]),
+ "paper_id": int(data[2]),
+ "faculty_id": int(data[0]),
+ }
+ if len(data) > 4:
+ retval["sid"] = data[4]
+
+ return retval
+
+ def get_paper_ocr_data(self):
+ data = self.get_code_data()
+ data["qr"] = self.QRData
+ data["errors"] = self.errors
+ data["warnings"] = self.warnings
+ data["up_position"] = (
+ list(self.xMarkerLocations[1] / self.imgWidth),
+ list(self.yMarkerLocations[1] / self.imgHeight),
+ )
+ data["right_position"] = (
+ list(self.xMarkerLocations[1] / self.imgWidth),
+ list(self.yMarkerLocations[1] / self.imgHeight),
+ )
+ data["ans_matrix"] = (
+ (np.array(self.answerMatrix) > self.settings["answer_threshold"]) * 1
+ ).tolist()
+ if data["sid"] is None and data["page_no"] == 0:
+ data["sid"] = self.get_enhanced_sid()
+ return data
--
Gitblit v1.9.3