278 lines
11 KiB
Plaintext
278 lines
11 KiB
Plaintext
{
|
|
"cells": [
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": 1,
|
|
"metadata": {},
|
|
"outputs": [],
|
|
"source": [
|
|
"import cv2\n",
|
|
"import matplotlib.pyplot as plt\n",
|
|
"import keras\n",
|
|
"import numpy as np\n",
|
|
"import threading\n",
|
|
"import tensorflow as tf"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": 2,
|
|
"metadata": {},
|
|
"outputs": [],
|
|
"source": [
|
|
"def wrap_frozen_graph(graph_def, inputs, outputs, print_graph=False):\n",
|
|
" def _imports_graph_def():\n",
|
|
" tf.compat.v1.import_graph_def(graph_def, name=\"\")\n",
|
|
"\n",
|
|
" wrapped_import = tf.compat.v1.wrap_function(_imports_graph_def, [])\n",
|
|
" import_graph = wrapped_import.graph\n",
|
|
"\n",
|
|
" if print_graph == True:\n",
|
|
" print(\"-\" * 50)\n",
|
|
" print(\"Frozen model layers: \")\n",
|
|
" layers = [op.name for op in import_graph.get_operations()]\n",
|
|
" for layer in layers:\n",
|
|
" print(layer)\n",
|
|
" print(\"-\" * 50)\n",
|
|
"\n",
|
|
" return wrapped_import.prune(\n",
|
|
" tf.nest.map_structure(import_graph.as_graph_element, inputs),\n",
|
|
" tf.nest.map_structure(import_graph.as_graph_element, outputs))"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": 3,
|
|
"metadata": {},
|
|
"outputs": [
|
|
{
|
|
"name": "stdout",
|
|
"output_type": "stream",
|
|
"text": [
|
|
"--------------------------------------------------\n",
|
|
"Frozen model layers: \n",
|
|
"x\n",
|
|
"sequential/conv2d/Conv2D/ReadVariableOp/resource\n",
|
|
"sequential/conv2d/Conv2D/ReadVariableOp\n",
|
|
"sequential/conv2d/Conv2D\n",
|
|
"sequential/conv2d/BiasAdd/ReadVariableOp/resource\n",
|
|
"sequential/conv2d/BiasAdd/ReadVariableOp\n",
|
|
"sequential/conv2d/BiasAdd\n",
|
|
"sequential/conv2d/Relu\n",
|
|
"sequential/batch_normalization/ReadVariableOp/resource\n",
|
|
"sequential/batch_normalization/ReadVariableOp\n",
|
|
"sequential/batch_normalization/ReadVariableOp_1/resource\n",
|
|
"sequential/batch_normalization/ReadVariableOp_1\n",
|
|
"sequential/batch_normalization/FusedBatchNormV3/ReadVariableOp/resource\n",
|
|
"sequential/batch_normalization/FusedBatchNormV3/ReadVariableOp\n",
|
|
"sequential/batch_normalization/FusedBatchNormV3/ReadVariableOp_1/resource\n",
|
|
"sequential/batch_normalization/FusedBatchNormV3/ReadVariableOp_1\n",
|
|
"sequential/batch_normalization/FusedBatchNormV3\n",
|
|
"sequential/max_pooling2d/MaxPool\n",
|
|
"sequential/conv2d_1/Conv2D/ReadVariableOp/resource\n",
|
|
"sequential/conv2d_1/Conv2D/ReadVariableOp\n",
|
|
"sequential/conv2d_1/Conv2D\n",
|
|
"sequential/conv2d_1/BiasAdd/ReadVariableOp/resource\n",
|
|
"sequential/conv2d_1/BiasAdd/ReadVariableOp\n",
|
|
"sequential/conv2d_1/BiasAdd\n",
|
|
"sequential/conv2d_1/Relu\n",
|
|
"sequential/batch_normalization_1/ReadVariableOp/resource\n",
|
|
"sequential/batch_normalization_1/ReadVariableOp\n",
|
|
"sequential/batch_normalization_1/ReadVariableOp_1/resource\n",
|
|
"sequential/batch_normalization_1/ReadVariableOp_1\n",
|
|
"sequential/batch_normalization_1/FusedBatchNormV3/ReadVariableOp/resource\n",
|
|
"sequential/batch_normalization_1/FusedBatchNormV3/ReadVariableOp\n",
|
|
"sequential/batch_normalization_1/FusedBatchNormV3/ReadVariableOp_1/resource\n",
|
|
"sequential/batch_normalization_1/FusedBatchNormV3/ReadVariableOp_1\n",
|
|
"sequential/batch_normalization_1/FusedBatchNormV3\n",
|
|
"sequential/max_pooling2d_1/MaxPool\n",
|
|
"sequential/conv2d_2/Conv2D/ReadVariableOp/resource\n",
|
|
"sequential/conv2d_2/Conv2D/ReadVariableOp\n",
|
|
"sequential/conv2d_2/Conv2D\n",
|
|
"sequential/conv2d_2/BiasAdd/ReadVariableOp/resource\n",
|
|
"sequential/conv2d_2/BiasAdd/ReadVariableOp\n",
|
|
"sequential/conv2d_2/BiasAdd\n",
|
|
"sequential/conv2d_2/Relu\n",
|
|
"sequential/batch_normalization_2/ReadVariableOp/resource\n",
|
|
"sequential/batch_normalization_2/ReadVariableOp\n",
|
|
"sequential/batch_normalization_2/ReadVariableOp_1/resource\n",
|
|
"sequential/batch_normalization_2/ReadVariableOp_1\n",
|
|
"sequential/batch_normalization_2/FusedBatchNormV3/ReadVariableOp/resource\n",
|
|
"sequential/batch_normalization_2/FusedBatchNormV3/ReadVariableOp\n",
|
|
"sequential/batch_normalization_2/FusedBatchNormV3/ReadVariableOp_1/resource\n",
|
|
"sequential/batch_normalization_2/FusedBatchNormV3/ReadVariableOp_1\n",
|
|
"sequential/batch_normalization_2/FusedBatchNormV3\n",
|
|
"sequential/conv2d_3/Conv2D/ReadVariableOp/resource\n",
|
|
"sequential/conv2d_3/Conv2D/ReadVariableOp\n",
|
|
"sequential/conv2d_3/Conv2D\n",
|
|
"sequential/conv2d_3/BiasAdd/ReadVariableOp/resource\n",
|
|
"sequential/conv2d_3/BiasAdd/ReadVariableOp\n",
|
|
"sequential/conv2d_3/BiasAdd\n",
|
|
"sequential/conv2d_3/Relu\n",
|
|
"sequential/batch_normalization_3/ReadVariableOp/resource\n",
|
|
"sequential/batch_normalization_3/ReadVariableOp\n",
|
|
"sequential/batch_normalization_3/ReadVariableOp_1/resource\n",
|
|
"sequential/batch_normalization_3/ReadVariableOp_1\n",
|
|
"sequential/batch_normalization_3/FusedBatchNormV3/ReadVariableOp/resource\n",
|
|
"sequential/batch_normalization_3/FusedBatchNormV3/ReadVariableOp\n",
|
|
"sequential/batch_normalization_3/FusedBatchNormV3/ReadVariableOp_1/resource\n",
|
|
"sequential/batch_normalization_3/FusedBatchNormV3/ReadVariableOp_1\n",
|
|
"sequential/batch_normalization_3/FusedBatchNormV3\n",
|
|
"sequential/conv2d_4/Conv2D/ReadVariableOp/resource\n",
|
|
"sequential/conv2d_4/Conv2D/ReadVariableOp\n",
|
|
"sequential/conv2d_4/Conv2D\n",
|
|
"sequential/conv2d_4/BiasAdd/ReadVariableOp/resource\n",
|
|
"sequential/conv2d_4/BiasAdd/ReadVariableOp\n",
|
|
"sequential/conv2d_4/BiasAdd\n",
|
|
"sequential/conv2d_4/Relu\n",
|
|
"sequential/batch_normalization_4/ReadVariableOp/resource\n",
|
|
"sequential/batch_normalization_4/ReadVariableOp\n",
|
|
"sequential/batch_normalization_4/ReadVariableOp_1/resource\n",
|
|
"sequential/batch_normalization_4/ReadVariableOp_1\n",
|
|
"sequential/batch_normalization_4/FusedBatchNormV3/ReadVariableOp/resource\n",
|
|
"sequential/batch_normalization_4/FusedBatchNormV3/ReadVariableOp\n",
|
|
"sequential/batch_normalization_4/FusedBatchNormV3/ReadVariableOp_1/resource\n",
|
|
"sequential/batch_normalization_4/FusedBatchNormV3/ReadVariableOp_1\n",
|
|
"sequential/batch_normalization_4/FusedBatchNormV3\n",
|
|
"sequential/max_pooling2d_2/MaxPool\n",
|
|
"sequential/flatten/Const\n",
|
|
"sequential/flatten/Reshape\n",
|
|
"sequential/dense/MatMul/ReadVariableOp/resource\n",
|
|
"sequential/dense/MatMul/ReadVariableOp\n",
|
|
"sequential/dense/MatMul\n",
|
|
"sequential/dense/BiasAdd/ReadVariableOp/resource\n",
|
|
"sequential/dense/BiasAdd/ReadVariableOp\n",
|
|
"sequential/dense/BiasAdd\n",
|
|
"sequential/dense/Relu\n",
|
|
"sequential/dense_1/MatMul/ReadVariableOp/resource\n",
|
|
"sequential/dense_1/MatMul/ReadVariableOp\n",
|
|
"sequential/dense_1/MatMul\n",
|
|
"sequential/dense_1/BiasAdd/ReadVariableOp/resource\n",
|
|
"sequential/dense_1/BiasAdd/ReadVariableOp\n",
|
|
"sequential/dense_1/BiasAdd\n",
|
|
"sequential/dense_1/Relu\n",
|
|
"sequential/dense_2/MatMul/ReadVariableOp/resource\n",
|
|
"sequential/dense_2/MatMul/ReadVariableOp\n",
|
|
"sequential/dense_2/MatMul\n",
|
|
"sequential/dense_2/BiasAdd/ReadVariableOp/resource\n",
|
|
"sequential/dense_2/BiasAdd/ReadVariableOp\n",
|
|
"sequential/dense_2/BiasAdd\n",
|
|
"sequential/dense_2/Softmax\n",
|
|
"NoOp\n",
|
|
"Identity\n",
|
|
"--------------------------------------------------\n"
|
|
]
|
|
}
|
|
],
|
|
"source": [
|
|
" # Load frozen graph using TensorFlow 1.x functions\n",
|
|
"with tf.io.gfile.GFile(\"./frozen_models/frozen_graph2.pb\", \"rb\") as f:\n",
|
|
" graph_def = tf.compat.v1.GraphDef()\n",
|
|
" loaded = graph_def.ParseFromString(f.read())\n",
|
|
"\n",
|
|
"# Wrap frozen graph to ConcreteFunctions\n",
|
|
"frozen_func = wrap_frozen_graph(graph_def=graph_def,\n",
|
|
" inputs=[\"x:0\"],\n",
|
|
" outputs=[\"Identity:0\"],\n",
|
|
" print_graph=False)"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": 4,
|
|
"metadata": {},
|
|
"outputs": [],
|
|
"source": [
|
|
"cap = cv2.VideoCapture(\"rybki.mp4\")\n",
|
|
"cap.set(cv2.CAP_PROP_FPS, 60)\n",
|
|
"\n",
|
|
"class_names=sorted(['Fish', \"Jellyfish\", 'Penguin', 'Puffin', 'Shark', 'Starfish', 'Stingray'])\n",
|
|
"object_detector = cv2.createBackgroundSubtractorMOG2(history=100, varThreshold=50)\n",
|
|
"\n",
|
|
"\n",
|
|
"width = int(cap.get(cv2.CAP_PROP_FRAME_WIDTH)) \n",
|
|
"height = int(cap.get(cv2.CAP_PROP_FRAME_HEIGHT))\n",
|
|
"fps = cap.get(cv2.CAP_PROP_FPS)\n",
|
|
"out = cv2.VideoWriter('track_fish.avi', cv2.VideoWriter_fourcc(*'MJPG'), fps, (width, height))\n",
|
|
"\n",
|
|
"while True:\n",
|
|
" ret, frame = cap.read()\n",
|
|
" if(frame is not None):\n",
|
|
" roi = frame[100: 900,330:1900]\n",
|
|
" mask = object_detector.apply(roi)\n",
|
|
" _, mask = cv2.threshold(mask,254,255, cv2.THRESH_BINARY)\n",
|
|
" conturs, _ =cv2.findContours(mask, cv2.RETR_TREE, cv2.CHAIN_APPROX_SIMPLE)\n",
|
|
"\n",
|
|
" images = []\n",
|
|
"\n",
|
|
" for cnt in conturs:\n",
|
|
" area = cv2.contourArea(cnt)\n",
|
|
" if area > 200:\n",
|
|
" #cv2.drawContours(roi,[cnt],-1,(0,255,0),2)\n",
|
|
" x,y,w,h = cv2.boundingRect(cnt)\n",
|
|
" rectangle = cv2.rectangle(roi,(x,y),(x+w,y+h),(0,255,0),3)\n",
|
|
" image_to_predict = roi[y:y+h,x:x+w]\n",
|
|
" image_to_predict = cv2.resize(image_to_predict,(227,227))\n",
|
|
" # images.append((x,y,rectangle,np.expand_dims(image_to_predict,axis=0)))\n",
|
|
" \n",
|
|
" pred = frozen_func(x=tf.convert_to_tensor(image_to_predict[None, :], dtype='float32'))\n",
|
|
" label = class_names[np.argmax(pred)]\n",
|
|
" cv2.putText(rectangle, label, (x, y-10), cv2.FONT_HERSHEY_SIMPLEX, 0.9, (36,255,12), 1)\n",
|
|
" # if images:\n",
|
|
" # pred = model.predict(np.vstack([image[3] for image in images]))\n",
|
|
" # labels = [class_names[np.argmax(pre)] for pre in pred]\n",
|
|
" # for i,image in enumerate(images):\n",
|
|
" roi = cv2.resize(roi, (960, 540)) \n",
|
|
" cv2.imshow(\"roi\", roi)\n",
|
|
"\n",
|
|
" key = cv2.waitKey(30)\n",
|
|
" if key == 27:\n",
|
|
" break\n",
|
|
"\n",
|
|
" #out.write(frame)\n",
|
|
" else:\n",
|
|
" break\n",
|
|
"\n",
|
|
"\n",
|
|
"#out.release()\n",
|
|
"cap.release()\n",
|
|
"cv2.destroyAllWindows()"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": 59,
|
|
"metadata": {},
|
|
"outputs": [],
|
|
"source": []
|
|
}
|
|
],
|
|
"metadata": {
|
|
"kernelspec": {
|
|
"display_name": "um",
|
|
"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": "3.9.15"
|
|
},
|
|
"orig_nbformat": 4,
|
|
"vscode": {
|
|
"interpreter": {
|
|
"hash": "876e189cbbe99a9a838ece62aae1013186c4bb7e0254a10cfa2f9b2381853efb"
|
|
}
|
|
}
|
|
},
|
|
"nbformat": 4,
|
|
"nbformat_minor": 2
|
|
}
|