camera

billetes.ipynb

@@ -2,7 +2,7 @@
  "cells": [
   {
    "cell_type": "code",
-   "execution_count": 7,
+   "execution_count": 2,
    "metadata": {},
    "outputs": [
     {
@@ -34,7 +34,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 8,
+   "execution_count": 3,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -49,7 +49,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 9,
+   "execution_count": 4,
    "metadata": {},
    "outputs": [
     {
@@ -85,7 +85,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 10,
+   "execution_count": 5,
    "metadata": {},
    "outputs": [
     {
@@ -105,7 +105,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 11,
+   "execution_count": 6,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -134,7 +134,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 12,
+   "execution_count": 7,
    "metadata": {},
    "outputs": [
     {
@@ -182,7 +182,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 13,
+   "execution_count": 8,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -200,11 +200,22 @@
   },
   {
    "cell_type": "code",
-   "execution_count": null,
+   "execution_count": 9,
    "metadata": {},
-   "outputs": [],
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "<tensorflow.python.training.tracking.util.CheckpointLoadStatus at 0x1f2f632d0d0>"
+      ]
+     },
+     "execution_count": 9,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
    "source": [
-    "model.load_weights((weights_path+'/weights'))"
+    "model.load_weights((weights_path+'/weights2'))"
    ]
   },
   {
@@ -305,14 +316,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": null,
-   "metadata": {},
-   "outputs": [],
-   "source": []
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 15,
+   "execution_count": 19,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -328,7 +332,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 16,
+   "execution_count": 18,
    "metadata": {},
    "outputs": [
     {
@@ -341,10 +345,10 @@
     {
      "data": {
       "text/plain": [
-       "<matplotlib.image.AxesImage at 0x18999edd1c0>"
+       "<matplotlib.image.AxesImage at 0x1f2f8294910>"
       ]
      },
-     "execution_count": 16,
+     "execution_count": 18,
      "metadata": {},
      "output_type": "execute_result"
     },
@@ -371,7 +375,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 17,
+   "execution_count": 15,
    "metadata": {},
    "outputs": [
     {
@@ -380,7 +384,7 @@
        "'50'"
       ]
      },
-     "execution_count": 17,
+     "execution_count": 15,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -393,50 +397,94 @@
    ]
   },
   {
-   "cell_type": "markdown",
+   "cell_type": "code",
+   "execution_count": 22,
    "metadata": {},
+   "outputs": [],
    "source": [
-    "# camera input"
+    "def get_prediction(img):\n",
+    "    img =  np.array(img)/255.0\n",
+    "    #print(img.shape)\n",
+    "    #plt.imshow(img)\n",
+    "    result = model.predict(img[np.newaxis,...])\n",
+    "    result.shape\n",
+    "    predicted_class = tf.math.argmax(result[0], axis=-1)\n",
+    "    return class_names[int(predicted_class)]"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 18,
+   "execution_count": 20,
    "metadata": {},
    "outputs": [
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "100\n"
+      "(224, 224, 3)\n"
      ]
+    },
+    {
+     "data": {
+      "text/plain": [
+       "'50'"
+      ]
+     },
+     "execution_count": 20,
+     "metadata": {},
+     "output_type": "execute_result"
     }
    ],
    "source": [
-    "from IPython.display import clear_output\n",
+    "img = Image.open('test.jpg').resize((224,224))\n",
     "\n",
+    "get_prediction(img)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "# camera input"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 26,
+   "metadata": {},
+   "outputs": [],
+   "source": [
     "import cv2\n",
-    "captura = cv2.VideoCapture(0)\n",
-    "captura.set(cv2.CAP_PROP_FRAME_WIDTH, 951)\n",
-    "captura.set(cv2.CAP_PROP_FRAME_HEIGHT, 524)\n",
-    "while (captura.isOpened()):\n",
-    "  ret, imagen = captura.read()\n",
-    "  if ret == True:\n",
-    "   # color_coverted = cv2.cvtColor(imagen, cv2.COLOR_BGR2RGB)\n",
-    "    img=Image.fromarray(imagen)\n",
     "\n",
-    "    cv2.imshow('video', imagen)\n",
-    "    img =  cv2.resize(imagen, dsize=(224, 224), interpolation=cv2.INTER_CUBIC)\n",
-    "    img =  np.array(img)/255.0\n",
-    "    result = model.predict(img[np.newaxis,...])\n",
-    "    predicted_class = tf.math.argmax(result[0], axis=-1)\n",
-    "    \n",
-    "    clear_output()\n",
-    "    print(class_names[int(predicted_class)])\n",
-    "    if cv2.waitKey(1) & 0xFF == ord('s'):\n",
-    "      break\n",
-    "  else: break\n",
-    "captura.release()\n",
+    "\n",
+    "def funcion(img):\n",
+    "  gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)\n",
+    "  gray = cv2.GaussianBlur(gray, (21, 21), 0)\n",
+    "\n",
+    "  ret, imgt = cv2.threshold(gray, 138, 255, cv2.THRESH_BINARY_INV)\n",
+    "\n",
+    "  cv2.imshow(\"Image threshold\", imgt)\n",
+    "  countours, hierarchy = cv2.findContours(imgt.copy(), cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_SIMPLE)\n",
+    "  rectangles = [cv2.boundingRect(countour) for countour in countours]\n",
+    "  c = 0\n",
+    "  for i , rect in enumerate(rectangles):\n",
+    "    if rect[2] > 100 and rect[3] > 100:\n",
+    "      imgn = img[rect[1]:rect[1] + rect[3], rect[0]:rect[0] + rect[2]]\n",
+    "      imgn = cv2.resize(imgn, (100, 100))\n",
+    "      c += 1  \n",
+    "\n",
+    "      cv2.rectangle(img, (rect[0], rect[1]), (rect[0] + rect[2], rect[1] + rect[3]), (255, 0, 0), 2)\n",
+    "      cv2.putText(img, str(get_prediction(imgn)), (rect[0], rect[1]), cv2.FONT_HERSHEY_SIMPLEX, 1, (200, 0, 0), 3, cv2.LINE_AA)\n",
+    "  return img\n",
+    "\n",
+    "cam = cv2.VideoCapture(0)\n",
+    "while True:\n",
+    "    val, img = cam.read()\n",
+    "    img = funcion(img)\n",
+    "    cv2.imshow(\"Image funcion\",img)\n",
+    "    if cv2.waitKey(1) & 0xFF == ord('q'):\n",
+    "        break\n",
+    "cam.release()\n",
     "cv2.destroyAllWindows()"
    ]
   },