detecting digits

.idea/workspace.xml

@@ -19,12 +19,17 @@
     <select />
   </component>
   <component name="ChangeListManager">
-    <list default="true" id="828778c9-9d97-422f-a727-18ddbd059b85" name="Default Changelist" comment="going to pytorch on conda eve">
+    <list default="true" id="828778c9-9d97-422f-a727-18ddbd059b85" name="Default Changelist" comment="add coder.py">
-      <change afterPath="$PROJECT_DIR$/coder/coder.py" afterDir="false" />
+      <change afterPath="$PROJECT_DIR$/coder/12345.png" afterDir="false" />
-      <change afterPath="$PROJECT_DIR$/coder/digit_reco_model.pt" afterDir="false" />
+      <change afterPath="$PROJECT_DIR$/coder/ll.png" afterDir="false" />
+      <change afterPath="$PROJECT_DIR$/coder/testno.png" afterDir="false" />
       <change beforePath="$PROJECT_DIR$/.idea/workspace.xml" beforeDir="false" afterPath="$PROJECT_DIR$/.idea/workspace.xml" afterDir="false" />
+      <change beforePath="$PROJECT_DIR$/coder/coder.py" beforeDir="false" afterPath="$PROJECT_DIR$/coder/coder.py" afterDir="false" />
+      <change beforePath="$PROJECT_DIR$/coder/digit_reco_model.pt" beforeDir="false" afterPath="$PROJECT_DIR$/coder/digit_reco_model.pt" afterDir="false" />
+      <change beforePath="$PROJECT_DIR$/coder/digit_reco_model2.pt" beforeDir="false" afterPath="$PROJECT_DIR$/coder/digit_reco_model2.pt" afterDir="false" />
       <change beforePath="$PROJECT_DIR$/coder/digits_recognizer.py" beforeDir="false" afterPath="$PROJECT_DIR$/coder/digits_recognizer.py" afterDir="false" />
-      <change beforePath="$PROJECT_DIR$/coder/gr_test.png" beforeDir="false" />
+      <change beforePath="$PROJECT_DIR$/coder/image.py" beforeDir="false" afterPath="$PROJECT_DIR$/coder/image.py" afterDir="false" />
+      <change beforePath="$PROJECT_DIR$/coder/rocognizer.py" beforeDir="false" afterPath="$PROJECT_DIR$/coder/rocognizer.py" afterDir="false" />
     </list>
     <option name="SHOW_DIALOG" value="false" />
     <option name="HIGHLIGHT_CONFLICTS" value="true" />
@@ -73,17 +78,17 @@
     <property name="RunOnceActivity.ShowReadmeOnStart" value="true" />
     <property name="SHARE_PROJECT_CONFIGURATION_FILES" value="true" />
     <property name="WebServerToolWindowFactoryState" value="false" />
-    <property name="last_opened_file_path" value="$PROJECT_DIR$/coder/dataset" />
+    <property name="last_opened_file_path" value="$PROJECT_DIR$/coder" />
     <property name="restartRequiresConfirmation" value="false" />
     <property name="settings.editor.selected.configurable" value="com.jetbrains.python.configuration.PyActiveSdkModuleConfigurable" />
   </component>
   <component name="RecentsManager">
     <key name="CopyFile.RECENT_KEYS">
-      <recent name="C:\Users\Pawel Lukaszewicz\PycharmProjects\AL-2020\coder\dataset" />
       <recent name="C:\Users\Pawel Lukaszewicz\PycharmProjects\AL-2020\coder" />
+      <recent name="C:\Users\Pawel Lukaszewicz\PycharmProjects\AL-2020\coder\dataset" />
     </key>
   </component>
-  <component name="RunManager" selected="Python.digits_recognizer">
+  <component name="RunManager" selected="Python.rocognizer">
     <configuration default="true" type="PythonConfigurationType" factoryName="Python">
       <module name="wozek" />
       <option name="INTERPRETER_OPTIONS" value="" />
@@ -110,6 +115,9 @@
       <module name="wozek" />
       <option name="INTERPRETER_OPTIONS" value="" />
       <option name="PARENT_ENVS" value="true" />
+      <envs>
+        <env name="PYTHONUNBUFFERED" value="1" />
+      </envs>
       <option name="SDK_HOME" value="" />
       <option name="WORKING_DIRECTORY" value="$PROJECT_DIR$/coder" />
       <option name="IS_MODULE_SDK" value="true" />
@@ -129,6 +137,9 @@
       <module name="wozek" />
       <option name="INTERPRETER_OPTIONS" value="" />
       <option name="PARENT_ENVS" value="true" />
+      <envs>
+        <env name="PYTHONUNBUFFERED" value="1" />
+      </envs>
       <option name="SDK_HOME" value="" />
       <option name="WORKING_DIRECTORY" value="$PROJECT_DIR$/coder" />
       <option name="IS_MODULE_SDK" value="true" />
@@ -185,7 +196,7 @@
       <option name="INPUT_FILE" value="" />
       <method v="2" />
     </configuration>
-    <configuration name="train_nn" type="PythonConfigurationType" factoryName="Python" temporary="true" nameIsGenerated="true">
+    <configuration name="z8" type="PythonConfigurationType" factoryName="Python" temporary="true" nameIsGenerated="true">
       <module name="wozek" />
       <option name="INTERPRETER_OPTIONS" value="" />
       <option name="PARENT_ENVS" value="true" />
@@ -198,7 +209,7 @@
       <option name="ADD_CONTENT_ROOTS" value="true" />
       <option name="ADD_SOURCE_ROOTS" value="true" />
       <EXTENSION ID="PythonCoverageRunConfigurationExtension" runner="coverage.py" />
-      <option name="SCRIPT_NAME" value="C:\Users\Pawel Lukaszewicz\PycharmProjects\AL-2020\coder\train_nn.py" />
+      <option name="SCRIPT_NAME" value="$PROJECT_DIR$/coder/z8.py" />
       <option name="PARAMETERS" value="" />
       <option name="SHOW_COMMAND_LINE" value="true" />
       <option name="EMULATE_TERMINAL" value="false" />
@@ -209,18 +220,18 @@
     </configuration>
     <list>
       <item itemvalue="Python.image" />
-      <item itemvalue="Python.train_nn" />
       <item itemvalue="Python.rocognizer" />
       <item itemvalue="Python.digits_recognizer" />
       <item itemvalue="Python.coder" />
+      <item itemvalue="Python.z8" />
     </list>
     <recent_temporary>
       <list>
-        <item itemvalue="Python.digits_recognizer" />
-        <item itemvalue="Python.coder" />
         <item itemvalue="Python.rocognizer" />
+        <item itemvalue="Python.coder" />
+        <item itemvalue="Python.digits_recognizer" />
+        <item itemvalue="Python.z8" />
         <item itemvalue="Python.image" />
-        <item itemvalue="Python.train_nn" />
       </list>
     </recent_temporary>
   </component>
@@ -258,7 +269,10 @@
       <workItem from="1590436739719" duration="6325000" />
       <workItem from="1590443664804" duration="2943000" />
       <workItem from="1590497613517" duration="6041000" />
-      <workItem from="1590518246722" duration="12460000" />
+      <workItem from="1590518246722" duration="12616000" />
+      <workItem from="1590559069326" duration="13892000" />
+      <workItem from="1590575699320" duration="792000" />
+      <workItem from="1590766924835" duration="4315000" />
     </task>
     <task id="LOCAL-00001" summary="create Shelf">
       <created>1589815443652</created>
@@ -344,7 +358,14 @@
       <option name="project" value="LOCAL" />
       <updated>1590447313737</updated>
     </task>
-    <option name="localTasksCounter" value="13" />
+    <task id="LOCAL-00013" summary="add coder.py">
+      <created>1590538529471</created>
+      <option name="number" value="00013" />
+      <option name="presentableId" value="LOCAL-00013" />
+      <option name="project" value="LOCAL" />
+      <updated>1590538529471</updated>
+    </task>
+    <option name="localTasksCounter" value="14" />
     <servers />
   </component>
   <component name="TypeScriptGeneratedFilesManager">
@@ -376,7 +397,8 @@
     <MESSAGE value="po" />
     <MESSAGE value="new dataset" />
     <MESSAGE value="going to pytorch on conda eve" />
-    <option name="LAST_COMMIT_MESSAGE" value="going to pytorch on conda eve" />
+    <MESSAGE value="add coder.py" />
+    <option name="LAST_COMMIT_MESSAGE" value="add coder.py" />
   </component>
   <component name="WindowStateProjectService">
     <state x="115" y="162" key="#com.intellij.refactoring.safeDelete.UnsafeUsagesDialog" timestamp="1589923610328">
@@ -403,10 +425,10 @@
       <screen x="0" y="0" width="1536" height="824" />
     </state>
     <state x="277" y="57" key="SettingsEditor/0.0.1536.824@0.0.1536.824" timestamp="1590443566792" />
-    <state x="361" y="145" key="Vcs.Push.Dialog.v2" timestamp="1590447321698">
+    <state x="361" y="145" key="Vcs.Push.Dialog.v2" timestamp="1590538563335">
       <screen x="0" y="0" width="1536" height="824" />
     </state>
-    <state x="361" y="145" key="Vcs.Push.Dialog.v2/0.0.1536.824@0.0.1536.824" timestamp="1590447321698" />
+    <state x="361" y="145" key="Vcs.Push.Dialog.v2/0.0.1536.824@0.0.1536.824" timestamp="1590538563335" />
     <state x="54" y="145" width="672" height="678" key="search.everywhere.popup" timestamp="1589918982407">
       <screen x="0" y="0" width="1536" height="824" />
     </state>

coder/coder.py

@@ -4,47 +4,49 @@ import torchvision
 import matplotlib.pyplot as plt
 from time import time
 from torchvision import datasets, transforms
-from torch import nn, optim, nn, optim
+from torch import nn, optim
 import cv2
 
 
-def view_classify(img, ps):
+transform = transforms.Compose([transforms.ToTensor(),
-    ''' Function for viewing an image and it's predicted classes.
+                                transforms.Normalize((0.5,), (0.5,)),
-    '''
+                                ])
-    ps = ps.data.numpy().squeeze()
+
 
-    fig, (ax1, ax2) = plt.subplots(figsize=(6,9), ncols=2)
-    ax1.imshow(img.resize_(1, 28, 28).numpy().squeeze())
-    ax1.axis('off')
-    ax2.barh(np.arange(10), ps)
-    ax2.set_aspect(0.1)
-    ax2.set_yticks(np.arange(10))
-    ax2.set_yticklabels(np.arange(10))
-    ax2.set_title('Class Probability')
-    ax2.set_xlim(0, 1.1)
-    plt.tight_layout()
 
 # load nn model
-model = torch.load('digit_reco_model2.pt')
+input_size = 784 # = 28*28
+hidden_sizes = [128, 128, 64]
+output_size = 10
+model = nn.Sequential(nn.Linear(input_size, hidden_sizes[0]),
+                      nn.ReLU(),
+                      nn.Linear(hidden_sizes[0], hidden_sizes[1]),
+                      nn.ReLU(),
+                      nn.Linear(hidden_sizes[1], hidden_sizes[2]),
+                      nn.ReLU(),
+                      nn.Linear(hidden_sizes[2], output_size),
+                      nn.LogSoftmax(dim=-1))
+model.load_state_dict(torch.load('digit_reco_model2.pt'))
+model.eval()
+# model = torch.load('digit_reco_model2.pt')
 
 if model is None:
     print("Model is not loaded.")
 else:
     print("Model is loaded.")
 
-# image
-img = cv2.cvtColor(cv2.imread('test3.png'), cv2.COLOR_BGR2GRAY)
-img = cv2.blur(img, (9, 9))  # poprawia jakosc
-img = cv2.resize(img, (28, 28), interpolation=cv2.INTER_AREA)
-img = img.reshape((len(img), -1))
 
-print(type(img))
+# img from dataset
-# print(img.shape)
+val_set = datasets.MNIST('PATH_TO_STORE_TESTSET', download=True, train=False, transform=transform)
-# plt.imshow(img ,cmap='binary')
+
-# plt.show()
+val_loader = torch.utils.data.DataLoader(val_set, batch_size=64, shuffle=True)
-img = np.array(img, dtype=np.float32)
+
-img = torch.from_numpy(img)
+images, labels = next(iter(val_loader))
-img = img.view(1, 784)
+print(type(images))
+img = images[0].view(1, 784)
+plt.imshow(images[0].numpy().squeeze(), cmap='gray_r')
+plt.show()
+
 
 # recognizing
 
@@ -54,5 +56,3 @@ with torch.no_grad():
 ps = torch.exp(logps)
 probab = list(ps.numpy()[0])
 print("Predicted Digit =", probab.index(max(probab)))
-
-view_classify(img.view(1, 28, 28), ps)

coder/digits_recognizer.py

@@ -5,6 +5,7 @@ import matplotlib.pyplot as plt
 from time import time
 from torchvision import datasets, transforms
 from torch import nn, optim
+import cv2
 
 # IMG transform
 transform = transforms.Compose([transforms.ToTensor(),
@@ -17,15 +18,6 @@ val_set = datasets.MNIST('PATH_TO_STORE_TESTSET', download=True, train=False, tr
 train_loader = torch.utils.data.DataLoader(train_set, batch_size=64, shuffle=True)
 val_loader = torch.utils.data.DataLoader(val_set, batch_size=64, shuffle=True)
 
-data_iter = iter(train_loader)
-images, labels = data_iter.next()
-
-print(images.shape)
-print(labels.shape)
-
-plt.imshow(images[0].numpy().squeeze(), cmap='gray_r')
-plt.show()
-
 # building nn model
 input_size = 784 # = 28*28
 hidden_sizes = [128, 128, 64]
@@ -48,15 +40,12 @@ images = images.view(images.shape[0], -1)
 logps = model(images)  # log probabilities
 loss = criterion(logps, labels)  # calculate the NLL loss
 
-# print('Before backward pass: \n', model[0].weight.grad)
-loss.backward()
-# print('After backward pass: \n', model[0].weight.grad)
 
 # training
 
 optimizer = optim.SGD(model.parameters(), lr=0.003, momentum=0.9)
 time0 = time()
-epochs = 100
+epochs = 1
 for e in range(epochs):
     running_loss = 0
     for images, labels in train_loader:
@@ -84,7 +73,6 @@ print("\nTraining Time (in minutes) =", (time() - time0) / 60)
 # testing
 
 images, labels = next(iter(val_loader))
-print(type(images))
 img = images[0].view(1, 784)
 with torch.no_grad():
     logps = model(img)
@@ -115,5 +103,5 @@ print("\nModel Accuracy =", (correct_count / all_count))
 
 # saving model
 
-# torch.save(model, './digit_reco_model.pt')
+# torch.save(model.state_dict(), './digit_reco_model.pt')
-torch.save(model, './digit_reco_model2.pt')
+# torch.save(model.state_dict(), './digit_reco_model2.pt')

coder/image.py

@@ -6,6 +6,7 @@ from sklearn.metrics import accuracy_score
 from sklearn.neural_network import MLPClassifier
 import pandas as pd
 import cv2
+import keras
 
 # 28x28
 train_data = np.genfromtxt('dataset/train.csv', delimiter=',', skip_header=1, max_rows=20000, encoding='utf-8')

coder/rocognizer.py

@@ -3,35 +3,34 @@ import argparse
 import imutils
 import cv2
 import matplotlib.pyplot as plt
+import torch
+from PIL import Image
 
-img = cv2.cvtColor(cv2.imread('barcode.jpg'), cv2.COLOR_BGR2GRAY)
+path = "test1.jpg"
 
-ddepth = cv2.cv.CV_32F if imutils.is_cv2() else cv2.CV_32F
+img = cv2.imread(path)
-X = cv2.Sobel(img, ddepth=ddepth, dx=1, dy=0, ksize=-1)
-Y = cv2.Sobel(img, ddepth=ddepth, dx=0, dy=1, ksize=-1)
 
-gradient = cv2.subtract(X, Y)
+img_gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
-gradient = cv2.convertScaleAbs(gradient)
+img_gray = cv2.GaussianBlur(img_gray, (5, 5), 0)
 
-blurred = cv2.blur(gradient, (9, 9))
+ret, im_th = cv2.threshold(img_gray, 90, 255, cv2.THRESH_BINARY_INV)
-(_, thresh) = cv2.threshold(blurred, 225, 255, cv2.THRESH_BINARY)
 
-kernel = cv2.getStructuringElement(cv2.MORPH_RECT, (21, 7))
+ctrs, hier = cv2.findContours(im_th.copy(), cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_SIMPLE)
-closed = cv2.morphologyEx(thresh, cv2.MORPH_CLOSE, kernel)
-closed = cv2.erode(closed, None, iterations=4)
-closed = cv2.dilate(closed, None, iterations=4)
 
-cnts = cv2.findContours(closed.copy(), cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_SIMPLE)
+rects = [cv2.boundingRect(ctr) for ctr in ctrs]
-cnts = imutils.grab_contours(cnts)
-c = sorted(cnts, key=cv2.contourArea, reverse=True)[0]
 
-rect = cv2.minAreaRect(c)
+for rect in rects:
-box = cv2.cv.BoxPoints(rect) if imutils.is_cv2() else cv2.boxPoints(rect)
+    # Draw the rectangles
-box = np.int0(box)
+    cv2.rectangle(img, (rect[0], rect[1]), (rect[0] + rect[2], rect[1] + rect[3]), (0, 255, 0), 3)
+    # Make the rectangular region around the digit
+    leng = int(rect[3] * 1.6)
+    pt1 = int(rect[1] + rect[3] // 2 - leng // 2)
+    pt2 = int(rect[0] + rect[2] // 2 - leng // 2)
+    roi = im_th[pt1:pt1+leng, pt2:pt2+leng]
+    # Resize the image
+    roi = cv2.resize(roi, (28, 28), interpolation=cv2.INTER_AREA)
+    roi = cv2.dilate(roi, (3, 3))
+    # Calculate the HOG features
 
-cv2.drawContours(img, [box], -1, (0, 255, 0), 3)
+cv2.imshow("Resulting Image with Rectangular ROIs", img)
-cv2.imshow("Image", img)
+cv2.waitKey()
-cv2.waitKey(0)
-
-plt.imshow(closed ,cmap='binary')
-plt.show()