F all in one.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "25629e3e",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[INFO] Loading images ...\n",
      "[INFO] Processing Healthy_Leaf ...\n",
      "[INFO] Processing Healthy_Nut ...\n",
      "[INFO] Processing Healthy_Trunk ...\n",
      "[INFO] Processing Mahali_Koleroga ...\n",
      "[INFO] Processing Stem_bleeding ...\n",
      "[INFO] Processing yellow_leaf spot_disease ...\n",
      "[INFO] Image loading completed\n",
      "['Healthy_Leaf' 'Healthy_Nut' 'Healthy_Trunk' 'Mahali_Koleroga'\n",
      " 'Stem_bleeding' 'yellow_leaf spot_disease']\n",
      "[INFO] Spliting data to train, test\n",
      "Model: \"sequential\"\n",
      "_________________________________________________________________\n",
      " Layer (type)                Output Shape              Param #   \n",
      "=================================================================\n",
      " conv2d (Conv2D)             (None, 254, 254, 32)      896       \n",
      "                                                                 \n",
      " max_pooling2d (MaxPooling2D  (None, 127, 127, 32)     0         \n",
      " )                                                               \n",
      "                                                                 \n",
      " conv2d_1 (Conv2D)           (None, 125, 125, 32)      9248      \n",
      "                                                                 \n",
      " max_pooling2d_1 (MaxPooling  (None, 62, 62, 32)       0         \n",
      " 2D)                                                             \n",
      "                                                                 \n",
      " flatten (Flatten)           (None, 123008)            0         \n",
      "                                                                 \n",
      " dense (Dense)               (None, 1000)              123009000 \n",
      "                                                                 \n",
      " dropout (Dropout)           (None, 1000)              0         \n",
      "                                                                 \n",
      " dense_1 (Dense)             (None, 500)               500500    \n",
      "                                                                 \n",
      " dropout_1 (Dropout)         (None, 500)               0         \n",
      "                                                                 \n",
      " dense_2 (Dense)             (None, 250)               125250    \n",
      "                                                                 \n",
      " dropout_2 (Dropout)         (None, 250)               0         \n",
      "                                                                 \n",
      " dense_3 (Dense)             (None, 125)               31375     \n",
      "                                                                 \n",
      " dense_4 (Dense)             (None, 6)                 756       \n",
      "                                                                 \n",
      "=================================================================\n",
      "Total params: 123,677,025\n",
      "Trainable params: 123,677,025\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n",
      "Epoch 1/20\n",
      "1/1 [==============================] - 34s 34s/step - loss: 1.9521 - accuracy: 0.1705 - val_loss: 4.5694 - val_accuracy: 0.3043\n",
      "Epoch 2/20\n",
      "1/1 [==============================] - 23s 23s/step - loss: 6.7931 - accuracy: 0.2841 - val_loss: 7.5651 - val_accuracy: 0.3043\n",
      "Epoch 3/20\n",
      "1/1 [==============================] - 13s 13s/step - loss: 8.8629 - accuracy: 0.2841 - val_loss: 4.2993 - val_accuracy: 0.3043\n",
      "Epoch 4/20\n",
      "1/1 [==============================] - 15s 15s/step - loss: 6.3728 - accuracy: 0.3409 - val_loss: 2.4797 - val_accuracy: 0.3043\n",
      "Epoch 5/20\n",
      "1/1 [==============================] - 15s 15s/step - loss: 5.6066 - accuracy: 0.2273 - val_loss: 2.1833 - val_accuracy: 0.3043\n",
      "Epoch 6/20\n",
      "1/1 [==============================] - 16s 16s/step - loss: 3.8829 - accuracy: 0.3068 - val_loss: 1.7765 - val_accuracy: 0.3043\n",
      "Epoch 7/20\n",
      "1/1 [==============================] - 15s 15s/step - loss: 2.7565 - accuracy: 0.3068 - val_loss: 1.5514 - val_accuracy: 0.3043\n",
      "Epoch 8/20\n",
      "1/1 [==============================] - 14s 14s/step - loss: 2.0444 - accuracy: 0.3636 - val_loss: 1.4891 - val_accuracy: 0.3043\n",
      "Epoch 9/20\n",
      "1/1 [==============================] - 16s 16s/step - loss: 1.6487 - accuracy: 0.3750 - val_loss: 1.4783 - val_accuracy: 0.4783\n",
      "Epoch 10/20\n",
      "1/1 [==============================] - 15s 15s/step - loss: 1.6181 - accuracy: 0.2386 - val_loss: 1.5462 - val_accuracy: 0.5652\n",
      "Epoch 11/20\n",
      "1/1 [==============================] - 13s 13s/step - loss: 1.5881 - accuracy: 0.3068 - val_loss: 1.5512 - val_accuracy: 0.5217\n",
      "Epoch 12/20\n",
      "1/1 [==============================] - 16s 16s/step - loss: 1.5043 - accuracy: 0.4205 - val_loss: 1.5363 - val_accuracy: 0.5652\n",
      "Epoch 13/20\n",
      "1/1 [==============================] - 16s 16s/step - loss: 1.4295 - accuracy: 0.4659 - val_loss: 1.5114 - val_accuracy: 0.5652\n",
      "Epoch 14/20\n",
      "1/1 [==============================] - 14s 14s/step - loss: 1.4858 - accuracy: 0.4318 - val_loss: 1.4392 - val_accuracy: 0.4783\n",
      "Epoch 15/20\n",
      "1/1 [==============================] - 15s 15s/step - loss: 1.4367 - accuracy: 0.4659 - val_loss: 1.3842 - val_accuracy: 0.3913\n",
      "Epoch 16/20\n",
      "1/1 [==============================] - 16s 16s/step - loss: 1.2795 - accuracy: 0.5341 - val_loss: 1.3329 - val_accuracy: 0.5217\n",
      "Epoch 17/20\n",
      "1/1 [==============================] - 16s 16s/step - loss: 1.2756 - accuracy: 0.5341 - val_loss: 1.2729 - val_accuracy: 0.5217\n",
      "Epoch 18/20\n",
      "1/1 [==============================] - 14s 14s/step - loss: 1.2292 - accuracy: 0.5114 - val_loss: 1.2026 - val_accuracy: 0.6522\n",
      "Epoch 19/20\n",
      "1/1 [==============================] - 9s 9s/step - loss: 1.2122 - accuracy: 0.5114 - val_loss: 1.1478 - val_accuracy: 0.7391\n",
      "Epoch 20/20\n",
      "1/1 [==============================] - 5s 5s/step - loss: 1.1605 - accuracy: 0.5682 - val_loss: 1.1242 - val_accuracy: 0.6087\n",
      "1/1 [==============================] - 1s 1s/step - loss: 1.1899 - accuracy: 0.5357\n"
     ]
    },
    {
     "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",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[INFO] Calculating model accuracy\n",
      "1/1 [==============================] - 1s 1s/step - loss: 1.1899 - accuracy: 0.5357\n",
      "Test Accuracy: 53.57142686843872\n",
      "1/1 [==============================] - 2s 2s/step\n",
      "[[2.1513805e-03 9.8303741e-01 3.8330127e-03 2.3752474e-03 8.4497184e-03\n",
      "  1.5322329e-04]]\n",
      "Healthy_Leaf 0.22 %\n",
      "Healthy_Nut 98.3 %\n",
      "Healthy_Trunk 0.38 %\n",
      "Mahali_Koleroga 0.24 %\n",
      "Stem_bleeding 0.84 %\n",
      "yellow_leaf spot_disease 0.02 %\n"
     ]
    }
   ],
   "source": [
    "import numpy as np\n",
    "import pickle\n",
    "import cv2\n",
    "from os import listdir\n",
    "from sklearn.preprocessing import LabelBinarizer\n",
    "from keras.models import Sequential\n",
    "from tensorflow.keras.layers import BatchNormalization\n",
    "from keras.layers.convolutional import Conv2D\n",
    "from keras.layers.convolutional import MaxPooling2D\n",
    "from keras.layers.core import Activation, Flatten, Dropout, Dense\n",
    "from keras import backend as K\n",
    "from keras.preprocessing.image import ImageDataGenerator\n",
    "from keras.optimizers import Adam\n",
    "from keras.preprocessing import image\n",
    "from tensorflow.keras.utils import img_to_array\n",
    "from sklearn.preprocessing import MultiLabelBinarizer\n",
    "from sklearn.model_selection import train_test_split\n",
    "import matplotlib.pyplot as plt\n",
    "import os\n",
    "import pathlib\n",
    "#1\n",
    "EPOCHS = 25\n",
    "INIT_LR = 1e-3\n",
    "BS = 32\n",
    "default_image_size = tuple((256, 256))\n",
    "image_size = 0\n",
    "path = pathlib.Path('C:\\Dataset')\n",
    "width=256\n",
    "height=256\n",
    "depth=3\n",
    "#2\n",
    "def convert_image_to_array(image_dir):\n",
    "    try:\n",
    "        image = cv2.imread(image_dir)\n",
    "        if image is not None :\n",
    "            image = cv2.resize(image, default_image_size)   \n",
    "            return img_to_array(image)\n",
    "        else :\n",
    "            return np.array([])\n",
    "    except Exception as e:\n",
    "        print(f\"Error : {e}\")\n",
    "        return None\n",
    "#3  \n",
    "image_list, label_list = [], []\n",
    "l=[]\n",
    "try:\n",
    "    print(\"[INFO] Loading images ...\")\n",
    "    root_dir = listdir(path)\n",
    "    for directory in root_dir :\n",
    "        # remove .DS_Store from list\n",
    "        if directory == \".DS_Store\" :\n",
    "            root_dir.remove(directory)\n",
    "\n",
    "    for plant_folder in root_dir :\n",
    "        plant_disease_folder_list = listdir(f\"{path}/{plant_folder}\")\n",
    "        \n",
    "        for disease_folder in plant_disease_folder_list :\n",
    "            # remove .DS_Store from list\n",
    "            if disease_folder == \".DS_Store\" :\n",
    "                plant_disease_folder_list.remove(disease_folder)\n",
    "\n",
    "        for plant_disease_folder in plant_disease_folder_list:\n",
    "            print(f\"[INFO] Processing {plant_disease_folder} ...\")\n",
    "            plant_disease_image_list = listdir(f\"{path}/{plant_folder}/{plant_disease_folder}/\")\n",
    "                \n",
    "            for single_plant_disease_image in plant_disease_image_list :\n",
    "                if single_plant_disease_image == \".DS_Store\" :\n",
    "                    plant_disease_image_list.remove(single_plant_disease_image)\n",
    "\n",
    "            for image in plant_disease_image_list:\n",
    "                l.append(image)\n",
    "                image_directory = f\"{path}/{plant_folder}/{plant_disease_folder}/{image}\"\n",
    "                if image_directory.endswith(\".jpg\") == True or image_directory.endswith(\".JPG\") == True:\n",
    "                    image_list.append(convert_image_to_array(image_directory))\n",
    "                    label_list.append(plant_disease_folder)\n",
    "    print(\"[INFO] Image loading completed\")  \n",
    "except Exception as e:\n",
    "    print(f\"Error : {e}\")\n",
    "#4\n",
    "image_size = len(image_list)\n",
    "\n",
    "#5\n",
    "label_binarizer = LabelBinarizer()\n",
    "image_labels = label_binarizer.fit_transform(label_list)\n",
    "#pickle.dump(label_binarizer,open('label_transform.pkl', 'wb'))\n",
    "n_classes = len(label_binarizer.classes_)\n",
    "\n",
    "#6\n",
    "print(label_binarizer.classes_)\n",
    "\n",
    "#7\n",
    "#print(image_labels)\n",
    "\n",
    "#8\n",
    "np_image_list = np.array(image_list, dtype=np.float16) / 225.0\n",
    "\n",
    "#9\n",
    "print(\"[INFO] Spliting data to train, test\")\n",
    "x_train, x_test, y_train, y_test = train_test_split(np_image_list, image_labels, test_size=0.2, random_state = 42) \n",
    "\n",
    "#10\n",
    "aug = ImageDataGenerator(\n",
    "    rotation_range=25, width_shift_range=0.1,\n",
    "    height_shift_range=0.1, shear_range=0.2, \n",
    "    zoom_range=0.2,horizontal_flip=True, \n",
    "    fill_mode=\"nearest\")\n",
    "\n",
    "#11\n",
    "inputShape = (height, width, depth)\n",
    "chanDim = -1\n",
    "if K.image_data_format() == \"channels_first\":\n",
    "    inputShape = (depth, height, width)\n",
    "    chanDim = 1\n",
    "\n",
    "#Create models architecture\n",
    "model = Sequential()\n",
    "\n",
    "#Add the first layer\n",
    "model.add( Conv2D(32, (3,3), activation='relu', input_shape=inputShape))\n",
    "\n",
    "#Add a pooling layer\n",
    "model.add(MaxPooling2D(pool_size=(2,2)))\n",
    "\n",
    "#Add another convolutional layer\n",
    "model.add( Conv2D(32, (3,3), activation='relu'))\n",
    "\n",
    "#Add another pooling layer\n",
    "model.add(MaxPooling2D(pool_size=(2,2)))\n",
    "\n",
    "#Add a flatenning layer\n",
    "model.add(Flatten())\n",
    "\n",
    "#Add a layer with 1000 neurons\n",
    "model.add(Dense(1000, activation='relu'))\n",
    "\n",
    "#Add a dropout layer\n",
    "model.add(Dropout(0.5))\n",
    "\n",
    "#Add a layer with 500 neurons\n",
    "model.add(Dense(500, activation='relu'))\n",
    "\n",
    "#Add a dropout layer\n",
    "model.add(Dropout(0.5))\n",
    "\n",
    "#Add a layer with 250 neurons\n",
    "model.add(Dense(250, activation='relu'))\n",
    "\n",
    "#Add a dropout layer\n",
    "model.add(Dropout(0.5))\n",
    "\n",
    "#Add a layer with 125 neurons\n",
    "model.add(Dense(125, activation='relu'))\n",
    "\n",
    "#Add a layer with 10 neurons\n",
    "model.add(Dense(6, activation='softmax'))\n",
    "\n",
    "#12...................................\n",
    "model.summary()\n",
    "\n",
    "#13\n",
    "model.compile(loss = 'categorical_crossentropy',\n",
    "              optimizer = 'adam', metrics = ['accuracy'])\n",
    "\n",
    "#14\n",
    "history = model.fit(x_train, y_train, batch_size= 256,\n",
    "                 epochs = 20, \n",
    "                 validation_split = 0.2)\n",
    "\n",
    "#15\n",
    "model.evaluate(x_test, y_test)[1]\n",
    "\n",
    "#16\n",
    "#pwd\n",
    "\n",
    "#17\n",
    "#Visualize model accuracy\n",
    "plt.plot(history.history['accuracy'])\n",
    "plt.plot(history.history['val_accuracy'])\n",
    "plt.title('Model Accuracy')\n",
    "plt.ylabel('Accuracy')\n",
    "plt.xlabel('Epoch')\n",
    "plt.legend(['Train', 'Val'], loc = 'upper left')\n",
    "plt.show()\n",
    "\n",
    "#18\n",
    "#Visualize model loss\n",
    "plt.plot(history.history['loss'])\n",
    "plt.plot(history.history['val_loss'])\n",
    "plt.title('Model Loss')\n",
    "plt.ylabel('Loss')\n",
    "plt.xlabel('Epoch')\n",
    "plt.legend(['Train', 'Val'], loc = 'upper right')\n",
    "plt.show()\n",
    "\n",
    "#19\n",
    "print(\"[INFO] Calculating model accuracy\")\n",
    "scores = model.evaluate(x_test, y_test)\n",
    "print(f\"Test Accuracy: {scores[1]*100}\")\n",
    "\n",
    "#20\n",
    "Remedy = ['','','','','','Disease can be controlled by spraying with one per cent Bordeaux mixture or Dithane M 45 @ 3 gram/Liter.']\n",
    "\n",
    "#21\n",
    "#image_dir=\"C:\\Dataset\\plant_folder\\Healthy_Leaf\\\\Healthy_leaf (2).jpg\"\n",
    "image_dir=\"C:\\Dataset\\plant_folder\\Healthy_Nut\\\\healthy_nut (8).jpg\"\n",
    "#image_dir=\"C:\\Dataset\\plant_folder\\Healthy_Trunk\\\\Healthy (17).jpg\"\n",
    "#image_dir=\"C:\\Dataset\\plant_folder\\Mahali_Koleroga\\\\diseases_nut (1).jpg\"\n",
    "#image_dir=\"C:\\Dataset\\plant_folder\\Stem_bleeding\\\\diseased (3).jpg\"\n",
    "#image_dir=\"C:\\Dataset\\plant_folder\\yellow_leaf spot_disease\\\\yellow_leaf_spot (11).jpeg\"\n",
    "#image_dir = \"C://Users//hp//Downloads//Ayurvedic+Healing_Betel+leaf+versus+tobacco.jpg\"\n",
    "im=convert_image_to_array(image_dir)\n",
    "np_image_li = np.array(im, dtype=np.float16) / 225.0\n",
    "npp_image = np.expand_dims(np_image_li, axis=0)\n",
    "\n",
    "#22\n",
    "result=model.predict(npp_image)\n",
    "print(result)\n",
    "\n",
    "#23\n",
    "#result=model.predict(npp_image)\n",
    "itemindex = np.where(result==np.max(result))\n",
    "for i in range(len(result[0])):\n",
    "    print(label_binarizer.classes_[i]+\" \"+str(round(result[0][i]*100,2))+ \" %\")\n",
    "if itemindex[1][0]>2:\n",
    "    print(\"\\nRemedy for the \"+ label_binarizer.classes_[itemindex[1][0]]+\" :\\n\",Remedy[itemindex[1][0]])\n",
    "    "
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.9.12"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}