膨胀

02_classification.ipynb

@@ -40,7 +40,7 @@
     "color_dict = {(0, 0, 255): \"yangeng\", (255, 0, 0): 'beijing',(0, 255, 0): \"zibian\"} # 颜色对应的类别\n",
     "# color_dict = {(0, 0, 255): \"yangeng\"}\n",
     "# color_dict = {(255, 0, 0): 'beijing'}\n",
-    "color_dict = {(0, 255, 0): \"zibian\"}\n",
+    "# color_dict = {(0, 255, 0): \"zibian\"}\n",
     "label_index = {\"yangeng\": 1, \"beijing\": 0, \"zibian\":2} # 类别对应的序号\n",
     "show_samples = False # 是否展示样本\n",
     "\n",

02_classification.py

@@ -14,7 +14,7 @@ color_dict = {(0, 0, 255): "yangeng", (255, 0, 0): 'beijing', (0, 255, 0): "zibi
 # color_dict = {(255, 0, 0): 'beijing'}
 # color_dict = {(0, 255, 0): "zibian"}
 label_index = {"yangeng": 1, "beijing": 0, "zibian": 2}  # 类别对应的序号
-show_samples = False  # 是否展示样本
+show_samples = True  # 是否展示样本
 
 # 定义一些训练量
 threshold = 5  # 正样本周围多大范围内的还算是正样本

04_multi_classification.ipynb

@@ -14,7 +14,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 44,
+   "execution_count": 17,
    "outputs": [],
    "source": [
     "import datetime\n",
@@ -25,7 +25,7 @@
     "import numpy as np\n",
     "import pickle\n",
     "from sklearn.tree import DecisionTreeClassifier\n",
-    "%matplotlib notebook\n",
+    "# %matplotlib notebook\n",
     "from main_test import virtual_main\n",
     "from models import AnonymousColorDetector\n",
     "from utils import lab_scatter"
@@ -39,11 +39,11 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 45,
+   "execution_count": 18,
    "outputs": [],
    "source": [
     "img_path = r\"C:\\Users\\FEIJINTI\\Desktop\\721\\zazhi\\Image_2022_0721_1351_38_946-002034.bmp\"\n",
-    "model_path = [\"models/beijing_dt_2022-07-21_16-44.model\", \"models/tobacco_dt_2022-07-21_16-30.model\",\"models/zibian_dt_2022-07-21_16-45.model\"]"
+    "model_path = [\"models/beijing_dt_2022-07-21_16-44.model\", \"models/tobacco_dt_2022-07-21_16-30.model\",\"models/zibian_dt_2022-07-21_16-45.model\",\"dt_2022-07-21_17-19.model\"]"
    ],
    "metadata": {
     "collapsed": false,
@@ -54,7 +54,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 46,
+   "execution_count": 19,
    "outputs": [],
    "source": [
     "img = cv2.imread(img_path)[:, :, ::-1]\n",
@@ -71,11 +71,12 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 47,
+   "execution_count": 20,
    "outputs": [],
    "source": [
     "t_detector = AnonymousColorDetector(file_path=model_path[1])\n",
-    "t_result = t_detector.predict(img).astype(np.uint8)"
+    "t_result = t_detector.predict(img).astype(np.uint8)\n",
+    "t_result = cv2.dilate(t_result, kernel = np.ones((3, 3), np.uint8))"
    ],
    "metadata": {
     "collapsed": false,
@@ -86,7 +87,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 48,
+   "execution_count": 21,
    "outputs": [],
    "source": [
     "z_detector = AnonymousColorDetector(file_path=model_path[2])\n",
@@ -101,10 +102,11 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 49,
+   "execution_count": 22,
    "outputs": [],
    "source": [
-    "result = 1 - (b_result | t_result | z_result)"
+    "s_detector = AnonymousColorDetector(file_path=model_path[3])\n",
+    "s_result = s_detector.predict(img).astype(np.uint8)"
    ],
    "metadata": {
     "collapsed": false,
@@ -115,7 +117,21 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 50,
+   "execution_count": 25,
+   "outputs": [],
+   "source": [
+    "result = 1 - (b_result | t_result)"
+   ],
+   "metadata": {
+    "collapsed": false,
+    "pycharm": {
+     "name": "#%%\n"
+    }
+   }
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 26,
    "outputs": [
     {
      "data": {
@@ -128,26 +144,20 @@
     {
      "data": {
       "text/plain": "<IPython.core.display.HTML object>",
-      "text/html": "<div id='71e0e832-f293-40bc-9913-301125493d2e'></div>"
+      "text/html": "<div id='45f22664-39a5-478c-8758-331c7e98de1a'></div>"
      },
      "metadata": {},
      "output_type": "display_data"
-    },
-    {
-     "data": {
-      "text/plain": "<matplotlib.image.AxesImage at 0x24e8faaa440>"
-     },
-     "execution_count": 50,
-     "metadata": {},
-     "output_type": "execute_result"
     }
    ],
    "source": [
-    "fig,axs = plt.subplots(4,1)\n",
+    "fig,axs = plt.subplots(5,1)\n",
     "axs[0].imshow(b_result)\n",
     "axs[1].imshow(t_result)\n",
     "axs[2].imshow(z_result)\n",
-    "axs[3].imshow(result)"
+    "axs[3].imshow(result)\n",
+    "axs[4].imshow(1-s_result)\n",
+    "plt.savefig(\"1.png\",dpi=900)"
    ],
    "metadata": {
     "collapsed": false,
@@ -158,7 +168,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 50,
+   "execution_count": 24,
    "outputs": [],
    "source": [],
    "metadata": {