Scribd
Upload
54 views11 pages

Computer Vision Basics

The document provides an overview of various image processing techniques, including object detection, image classification, segmentation, and face recognition. It includes code snippets for manipulating images using OpenCV and TensorFlow, demonstrating operations like grayscale conversion, resizing, normalization, and data augmentation. Additionally, it covers advanced topics such as Generative Adversarial Networks (GANs) and text extraction from images using Tesseract.

Uploaded by

369 369
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
54 views11 pages

Computer Vision Basics

The document provides an overview of various image processing techniques, including object detection, image classification, segmentation, and face recognition. It includes code snippets for manipulating images using OpenCV and TensorFlow, demonstrating operations like grayscale conversion, resizing, normalization, and data augmentation. Additionally, it covers advanced topics such as Generative Adversarial Networks (GANs) and text extraction from images using Tesseract.

Uploaded by

369 369
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd

Image Processing –

Manipulating and analyzing


images
Object Detection –
Identifying objects in
images
Image Classification –
Assigning labels to images
Segmentation – Dividing
images into meaningful
regions
Face Recognition –
Identifying faces in images

@Tajamulkhann
Convert images to grayscale
Resize images
Normalize pixel values
Apply filters (Gaussian Blur,
Edge Detection)

import cv2
import numpy as np
image = [Link]("[Link]")
# Load image
gray = [Link](image, cv2.COLOR_BGR2GRAY)
# Convert to grayscale
resized = [Link](gray, (128, 128))
# Resize image
normalized = resized / 255.0
# Normalize pixel values

@Tajamulkhann
Sobel & Canny Edge Detection
Blurring (Gaussian, Median,
Bilateral)

edges = [Link](gray, 100, 200)


# Canny Edge Detection
blurred = [Link](gray,
(5,5), 0) # Gaussian Blur

@Tajamulkhann
Rotation, Flipping, Zooming,
Shearing
Used to increase dataset
diversity in deep learning

from
[Link]
import ImageDataGenerator

datagen =
ImageDataGenerator(rotation_range=30,
horizontal_flip=True)
augmented_image =
datagen.random_transform(image)

@Tajamulkhann
Classifies images into categories

import tensorflow as tf

model = [Link]([
[Link].Conv2D(32, (3,3),
activation='relu', input_shape=(128,128,3)),
[Link].MaxPooling2D(2,2),
[Link](),
[Link](128, activation='relu'),
[Link](10, activation='softmax')
])

[Link](optimizer='adam',
loss='categorical_crossentropy', metrics=
['accuracy'])

@Tajamulkhann
SSD, Faster R-CNN
YOLO (You Only Look Once) –
Real-time object detection

import cv2

net =
[Link]("[Link]",
"[Link]")
layer_names = [Link]()
output_layers = [layer_names[i - 1]
for i in
[Link]()]

@Tajamulkhann
Haar Cascades – Pre-trained
classifiers for detecting
faces

face_cascade =
[Link]([Link]
s + "haarcascade_frontalface_default.xml")
faces =
face_cascade.detectMultiScale(gray, 1.1,
4)

for (x, y, w, h) in faces:


[Link](image, (x, y), (x + w, y
+ h), (255, 0, 0), 2)

@Tajamulkhann
Divides an image into
meaningful parts

import cv2

ret, thresh = [Link](gray, 127,


255, cv2.THRESH_BINARY)
contours, hierarchy =
[Link](thresh, cv2.RETR_TREE,
cv2.CHAIN_APPROX_SIMPLE)
[Link](image, contours, -1,
(0,255,0), 3)

@Tajamulkhann
Extracts text from images

import pytesseract

text =
pytesseract.image_to_string
(gray)
print(text)

@Tajamulkhann
(GANs, Autoencoders)
GANs (Generative
Adversarial Networks) –
Generate new images)

from [Link] import Dense,


LeakyReLU
from [Link] import
Sequential

generator = Sequential([
Dense(256, input_dim=100),
LeakyReLU(alpha=0.2),
Dense(512, activation='relu'),
Dense(1024, activation='relu'),
Dense(784, activation='tanh')
])

@Tajamulkhann
Follow for more!

You might also like