0624

.ipynb_checkpoints/Accuracy_Testing-checkpoint.ipynb

@@ -2,8 +2,10 @@
  "cells": [
   {
    "cell_type": "code",
-   "execution_count": 9,
-   "metadata": {},
+   "execution_count": 28,
+   "metadata": {
+    "scrolled": true
+   },
    "outputs": [
     {
      "name": "stdout",
@@ -49,6 +51,18 @@
      },
      "output_type": "display_data"
     },
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 432x288 with 1 Axes>"
+      ]
+     },
+     "metadata": {
+      "needs_background": "light"
+     },
+     "output_type": "display_data"
+    },
     {
      "data": {
       "image/png": "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\n",
@@ -164,13 +178,13 @@
     "            plate, rect, origin_plate,flg  =plate\n",
     "            img =cv2.resize(origin_plate,(300,100))\n",
     "            img = cv2.bitwise_not(img)\n",
-    "            ###車排上下定位\n",
+    "            ### 車排上下定位\n",
     "            image_rgb = fm.findContoursAndDrawBoundingBox(img)\n",
     "            image_rgb = cv2.bitwise_not(image_rgb)\n",
     "            plt.figure(\"Imgae\"+str(j))\n",
     "            plt.imshow(image_rgb)\n",
     "            plt.show\n",
-    "            ###車排左右定位\n",
+    "            ### 車排左右定位\n",
     "            image_rgb = cv2.bitwise_not(image_rgb)\n",
     "            image_rgb_rl = fv.finemappingVertical(image_rgb)\n",
     "            image_rgb_rl = cv2.bitwise_not(image_rgb_rl)\n",
@@ -182,7 +196,20 @@
     "            ### 車牌辨識\n",
     "            image_gray = cv2.cvtColor(image_rgb_rl,cv2.COLOR_RGB2GRAY)\n",
     "            val = segmentation.slidingWindowsEval(image_gray)\n",
-    "            refined,name,con,nums=val\n",
+    "            refined,name,con,nums,cut_data=val\n",
+    "            \n",
+    "            ### 分割機率圖\n",
+    "            p=cut_data[0]\n",
+    "            lmin=cut_data[1]\n",
+    "            x=np.zeros(114)\n",
+    "            \n",
+    "            for i in lmin:\n",
+    "                x[i]=p[i]\n",
+    "                \n",
+    "            \n",
+    "            plt.figure()\n",
+    "            plt.plot(x,'*',p)                     \n",
+    "            \n",
     "\n",
     "            ### 切割出的字元\n",
     "            for i,one in enumerate (refined):\n",
@@ -190,10 +217,10 @@
     "                plt.imshow(one)\n",
     "                plt.show\n",
     "\n",
-    "            ###辨識結果\n",
+    "            ### 辨識結果\n",
     "            print('車牌',name,'可信度',con,'字數',nums)\n",
     "            \n",
-    "            return True\n",
+    "            return cut_data\n",
     "        \n",
     "        \n",
     "LPR=MYSimpleRecognizePlate(img)\n",
@@ -216,17 +243,31 @@
     }
    ],
    "source": [
+    "## GPU 狀態\n",
     "import tensorflow as tf\n",
-    "# Creates a graph.\n",
-    "a = tf.constant([1.0, 2.0, 3.0, 4.0, 5.0, 6.0], shape=[2, 3], name='a')\n",
-    "b = tf.constant([1.0, 2.0, 3.0, 4.0, 5.0, 6.0], shape=[3, 2], name='b')\n",
-    "c = tf.matmul(a, b)\n",
-    "# Creates a session with log_device_placement set to True.\n",
     "sess = tf.Session(config=tf.ConfigProto(log_device_placement=True))\n",
-    "# Runs the op.\n",
-    "print(sess.run(c))\n"
+    "\n"
    ]
   },
+  {
+   "cell_type": "code",
+   "execution_count": 13,
+   "metadata": {},
+   "outputs": [
+    {
+     "ename": "TypeError",
+     "evalue": "object of type 'numpy.float64' has no len()",
+     "output_type": "error",
+     "traceback": [
+      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
+      "\u001b[0;31mTypeError\u001b[0m                                 Traceback (most recent call last)",
+      "\u001b[0;32m<ipython-input-13-b9a0bc4b8794>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m\u001b[0m\n\u001b[0;32m----> 1\u001b[0;31m \u001b[0mlen\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mLPR\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;36m0\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;36m1\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m",
+      "\u001b[0;31mTypeError\u001b[0m: object of type 'numpy.float64' has no len()"
+     ]
+    }
+   ],
+   "source": []
+  },
   {
    "cell_type": "code",
    "execution_count": null,

Accuracy_Testing.ipynb

@@ -2,8 +2,10 @@
  "cells": [
   {
    "cell_type": "code",
-   "execution_count": 9,
-   "metadata": {},
+   "execution_count": 28,
+   "metadata": {
+    "scrolled": true
+   },
    "outputs": [
     {
      "name": "stdout",
@@ -49,6 +51,18 @@
      },
      "output_type": "display_data"
     },
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 432x288 with 1 Axes>"
+      ]
+     },
+     "metadata": {
+      "needs_background": "light"
+     },
+     "output_type": "display_data"
+    },
     {
      "data": {
       "image/png": "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\n",
@@ -164,13 +178,13 @@
     "            plate, rect, origin_plate,flg  =plate\n",
     "            img =cv2.resize(origin_plate,(300,100))\n",
     "            img = cv2.bitwise_not(img)\n",
-    "            ###車排上下定位\n",
+    "            ### 車排上下定位\n",
     "            image_rgb = fm.findContoursAndDrawBoundingBox(img)\n",
     "            image_rgb = cv2.bitwise_not(image_rgb)\n",
     "            plt.figure(\"Imgae\"+str(j))\n",
     "            plt.imshow(image_rgb)\n",
     "            plt.show\n",
-    "            ###車排左右定位\n",
+    "            ### 車排左右定位\n",
     "            image_rgb = cv2.bitwise_not(image_rgb)\n",
     "            image_rgb_rl = fv.finemappingVertical(image_rgb)\n",
     "            image_rgb_rl = cv2.bitwise_not(image_rgb_rl)\n",
@@ -182,7 +196,20 @@
     "            ### 車牌辨識\n",
     "            image_gray = cv2.cvtColor(image_rgb_rl,cv2.COLOR_RGB2GRAY)\n",
     "            val = segmentation.slidingWindowsEval(image_gray)\n",
-    "            refined,name,con,nums=val\n",
+    "            refined,name,con,nums,cut_data=val\n",
+    "            \n",
+    "            ### 分割機率圖\n",
+    "            p=cut_data[0]\n",
+    "            lmin=cut_data[1]\n",
+    "            x=np.zeros(114)\n",
+    "            \n",
+    "            for i in lmin:\n",
+    "                x[i]=p[i]\n",
+    "                \n",
+    "            \n",
+    "            plt.figure()\n",
+    "            plt.plot(x,'*',p)                     \n",
+    "            \n",
     "\n",
     "            ### 切割出的字元\n",
     "            for i,one in enumerate (refined):\n",
@@ -190,10 +217,10 @@
     "                plt.imshow(one)\n",
     "                plt.show\n",
     "\n",
-    "            ###辨識結果\n",
+    "            ### 辨識結果\n",
     "            print('車牌',name,'可信度',con,'字數',nums)\n",
     "            \n",
-    "            return True\n",
+    "            return cut_data\n",
     "        \n",
     "        \n",
     "LPR=MYSimpleRecognizePlate(img)\n",
@@ -222,6 +249,25 @@
     "\n"
    ]
   },
+  {
+   "cell_type": "code",
+   "execution_count": 13,
+   "metadata": {},
+   "outputs": [
+    {
+     "ename": "TypeError",
+     "evalue": "object of type 'numpy.float64' has no len()",
+     "output_type": "error",
+     "traceback": [
+      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
+      "\u001b[0;31mTypeError\u001b[0m                                 Traceback (most recent call last)",
+      "\u001b[0;32m<ipython-input-13-b9a0bc4b8794>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m\u001b[0m\n\u001b[0;32m----> 1\u001b[0;31m \u001b[0mlen\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mLPR\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;36m0\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;36m1\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m",
+      "\u001b[0;31mTypeError\u001b[0m: object of type 'numpy.float64' has no len()"
+     ]
+    }
+   ],
+   "source": []
+  },
   {
    "cell_type": "code",
    "execution_count": null,

hyperlpr/segmentation_0124.py

@@ -368,6 +368,8 @@ def slidingWindowsEval(image):
 #         cv2.imwrite("./mid_none/"+str(filename),image)
         return []
     pin = np.array(pin)
+    
+    cut_data=p,lmin
 
 #     nums=len(l.argrelmin(np.array(p),order = 3)[0])+1
 #    print("W:",nums)
@@ -451,4 +453,4 @@ def slidingWindowsEval(image):
     
 #    print("字符识别",time.time() - t0)
     
-    return refined,name,confidence,nums
+    return refined,name,confidence,nums,cut_data