Finding Data Points of an Image Using Harris Corner Detection Technique
Use Harris corner detection technique to find the data points (key points) of the image. Write a program to perform K-means clustering technique using the strongest 100 points of part 1. You need to create your own function/code for this part (K-mean algorithm).
• Compare the outcomes of multiple runs with different values for K and choose the best one based on a predefined criterion.
• Display the original image with the best detected clusters on it (use different colors).
• Draw a bounding box for each cluster of the data points of part 2.
You may use the built-in function/library (cv2.rectangle()) or any other one for this part. Display the original image with the bounding boxes on it (use different colors).
Cornerstone Detection code 1
{
"cells": [
{
"cell_type": "code",
"execution_count": 1,
"metadata": {},
"outputs": [],
"source": [
"importnumpy as np\n",
"import cv2 as cv\n",
"importmatplotlib.pyplot as plt"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"# Q1. Detect corner using Harris detector"
]
},
{
"cell_type": "code",
"execution_count": 2,
"metadata": {},
"outputs": [],
"source": [
"file = 'D:/freelancer/corner/image.jpg'"
]
},
{
"cell_type": "code",
"execution_count": 3,
"metadata": {},
"outputs": [],
"source": [
"im = cv.imread(file)"
]
},
{
"cell_type": "code",
"execution_count": 4,
"metadata": {},
"outputs": [],
"source": [
"gray = cv.cvtColor(im, cv.COLOR_BGR2GRAY)\n",
"gray = np.float32(gray)"
]
},
{
"cell_type": "code",
"execution_count": 5,
"metadata": {},
"outputs": [],
"source": [
"dst = cv.cornerHarris(gray, 2, 3, 0.04)"
]
},
{
"cell_type": "code",
"execution_count": 6,
"metadata": {},
"outputs": [],
"source": [
"im_new = im.copy()\n",
"im_new[dst>0.01*dst.max()]=[255,0,0]"
]
},
{
"cell_type": "code",
"execution_count": 7,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
""
]
},
"execution_count": 7,
"metadata": {},
"output_type": "execute_result"
},
{
"data": {
Cornerstone Code Solution 2
{
"cells": [
{
"cell_type": "code",
"execution_count": 250,
"metadata": {},
"outputs": [],
"source": [
"importnumpy as np\n",
"import cv2 as cv\n",
"importmatplotlib.pyplot as plt"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"# Detect corner using Harris detector"
]
},
{
"cell_type": "code",
"execution_count": 281,
"metadata": {},
"outputs": [],
"source": [
"file = 'D:/freelancer/corner/image.jpg'"
]
},
{
"cell_type": "code",
"execution_count": 282,
"metadata": {},
"outputs": [],
"source": [
"im = cv.imread(file)"
]
},
{
"cell_type": "code",
"execution_count": 283,
"metadata": {},
"outputs": [],
"source": [
"gray = cv.cvtColor(im, cv.COLOR_BGR2GRAY)\n",
"gray = np.float32(gray)"
]
},
{
"cell_type": "code",
"execution_count": 284,
"metadata": {},
"outputs": [],
"source": [
"dst = cv.cornerHarris(gray, 2, 3, 0.04)"
]
},
{
"cell_type": "code",
"execution_count": 285,
"metadata": {},
"outputs": [],
"source": [
"im_new = im.copy()\n",
"im_new[dst>0.01*dst.max()]=[255,0,0]"
]
},
{
"cell_type": "code",
"execution_count": 286,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
""
]
},
"execution_count": 286,
"metadata": {},
"output_type": "execute_result"
},
{
"data": {