Digital Image processing

class: center, middle, inverse, title-slide

.title[
# Digital Image processing
]
.subtitle[
## 🗺 <br /> .f3[~/Geospatial Techniques/Remote Sensing]
]
.author[
### <strong>Dr. Ankit Deshmukh</strong>
]
.institute[
### Pandit Deendayal Energy University, Gandhinagar
]
.date[
### 11 October 2022
]

---

class: center middle inverse

<div>
<style type="text/css">.xaringan-extra-logo {
width: 60px;
height: 60px;
z-index: 0;
background-image: url(./css/Anix-Logo.png);
background-size: contain;
background-repeat: no-repeat;
position: absolute;
top:1em;right:1em;
}
</style>
<script>(function () {
  let tries = 0
  function addLogo () {
    if (typeof slideshow === 'undefined') {
      tries += 1
      if (tries < 10) {
        setTimeout(addLogo, 100)
      }
    } else {
      document.querySelectorAll('.remark-slide-content:not(.title-slide):not(.inverse):not(.hide_logo)')
        .forEach(function (slide) {
          const logo = document.createElement('a')
          logo.classList = 'xaringan-extra-logo'
          logo.href = 'https://www.ankitdeshmukh.com/'
          slide.appendChild(logo)
        })
    }
  }
  document.addEventListener('DOMContentLoaded', addLogo)
})()</script>
</div><style>.xe__progress-bar__container {
  top:0;
  opacity: 1;
  position:absolute;
  right:0;
  left: 0;
}
.xe__progress-bar {
  height: 0.25em;
  background-color: #28282888;
  width: calc(var(--slide-current) / var(--slide-total) * 100%);
}
.remark-visible .xe__progress-bar {
  animation: xe__progress-bar__wipe 200ms forwards;
  animation-timing-function: cubic-bezier(.86,0,.07,1);
}
@keyframes xe__progress-bar__wipe {
  0% { width: calc(var(--slide-previous) / var(--slide-total) * 100%); }
  100% { width: calc(var(--slide-current) / var(--slide-total) * 100%); }
}</style>

# .white[Sooner or later all things are numbers, yes?] 
.b.center[~ Terry Pratchett ~]

---
# Topics to be covered for .blue["Digital Image processing"]

- Characteristics of a digital image

- Image enhancement

- Contrast manipulation

- Image registration

- Digital image interpretation techniques.

---
# Types of data products in remote sensing

- Depending on  correction applied  & level of processing
<img src="images/product.png" class="w-50 br4 dib center">

---
# Image interpretation strategy and elements of image interpretation
.pull-left[

- Visual image interpretation: Process of identifying what we see on the images and communicate the information obtained from these images to others for evaluating its significance.

- Includes relative locations and extents

]

.pull-right[
<img src="images/element.png" class="w-100 br4 dib center">
]

---
# Gray Scale
.pull-left[
- Most raw unprocessed satellite imagery is stored in a gray scale format.

- A gray scale is a color scale that ranges from black to white, with varying intermediate shades of gray.

- A commonly used gray scale for remote sensing image processing is a 256 shade gray scale.

- A value of 0 represents a pure black color, the value of 255 represents pure white, and each value in between represents a progressively darker shade of gray.

]

.pull-right[
<img src="images/Gray-Grid.png" class="w-100 br4 dib center">
]

---
# Multilayer Image
- Multilayer image is formed by "stacking“ images from the same area together.
- Each component image is a layer in the multilayer image and carry some specific information about the area.
- **Multilayer images can also be formed by combining images obtained from different sensors, and other subsidiary data.**

.footnote[https://semiautomaticclassificationmanual.readthedocs.io/en/latest/remote_sensing.html]

---
# One images with 3 Bands 
<img src="images/apple.png" class="w-100 br4 dib center">

---
# Three Broad Categories of Image Processing
.pull-left[
- Image Restoration (Pre-processing)
- Image Enhancement
- Classification and Information Extraction
]

.pull-right[
<img src="images/flow.png" class="w-80 br4 dib center">
]

---
# Image Restoration (Pre-processing)

- Radiometric and geometric corrections are required

- Atmospheric correction

- Sun illumination geometry

- Surface-induced geometric distortions

- Spacecraft velocity and attitude variations (roll, pitch, and yaw)

- Effects of Earth rotation, elevation, curvature (including skew effects),

- Abnormalities of instrument performance

- Loss of specific scan lines, etc.

---
# Image Enhancement
- **Contrast Enhancement**
<img src="images/Contrast.png" class="w-80 br4 dib center">

---
# A Histogram
- A histogram is an approximate representation of the distribution of numerical data. The term was first introduced by Karl Pearson.

- To assess the probability distribution of a given variable by depicting the frequencies of observations occurring in certain ranges of values.

.footnote[https://en.wikipedia.org/wiki/Histogram]

---
# Create a histogram with ramdom data
.pull-left[

```r
# How a create a histogram
sample <- c(2,3,1,4,2,2,4,4,4,2,2,1,4,5,3,2,1,6,3,3,3,1,1,1,1,2,2)
sort(sample)
```

```
##  [1] 1 1 1 1 1 1 1 2 2 2 2 2 2 2 2 3 3 3 3 3 4 4 4 4 4 5 6
```

```r
table(sample)
```

```
## sample
## 1 2 3 4 5 6 
## 7 8 5 5 1 1
```
## Roll a dice
https://www.calculator.net/dice-roller.html

]
.pull-right[

```r
barplot(table(sample))
```

```r
# Use inbuilt function in `r`
# hist(sample, n = length(table(sample)) + 1)
```
]

---
# Sampe image: An apple
.pull-left[
<img src="images/Apple-1.jpg" class="w-70 br4 dib center">
]
.pull-right[
<img src="images/Apple-2.jpg" class="w-70 br4 dib center">
]

---
# Plotting the bands
<img src="images/Hist3.png" class="w-70 br4 dib center">

---
# Plotting the bands

```r
if(!require(raster)){install.packages("raster");library(raster)}
png(filename = "images/Hist3.png", width = 1200, height = 850, res = 150)
Data <- brick("./images/Apple-1.jpg")
par(mfrow = c(2,3))
Band1 <- as.vector(Data$Apple.1.1)
hist(Band1[Band1 < 225], 50, main = "Red Band", col = "red")

Band2 <- as.vector(Data$Apple.1.2)
hist(Band2[Band2 < 255], 50, main = "Green Band", col = "green")

Band3 <- as.vector(Data$Apple.1.3)
hist(Band2[Band2 < 150], 50, main = "Blue Band", col = "blue")

Data <- brick("./images/Apple-2.jpg")
Band1 <- as.vector(Data$Apple.2.1)
hist(Band1[Band1 < 225], 50, main = NULL, col = "red")

Band2 <- as.vector(Data$Apple.2.2)
hist(Band2[Band2 < 255], 50, main = NULL, col = "green")

Band3 <- as.vector(Data$Apple.2.3)
hist(Band2[Band2 < 150], 50, main = NULL, col = "blue")
dev.off()
```

---
# Linear Contrast Enhancement
<img src="images/Li-1.png" class="w-40 br1 dib center">
Displayed in an 8-bit system. DN values range from 60-158. DN values from 0-59 and 159-255 are not utilized.

<img src="images/Li-2.png" class="w-40 br1 dib center">
DN values are stretched to 0-255, Contrast is improved, Light tones appear lighter, Dark tones appear darker.

---
# Non-linear Contrast Stretching
DN values are not stretched linearly to uniformly occupy the entire display range, example of this is **"Histogram-equalized stretch"**

---
# Histogram-equalized stretch
DN values are enhanced based on their frequency in the original image.
Example: Enhancement in an 8-bit display system

### `$$DN_{st} = \sum_{j=1}^k \frac{n_j}{N}$$`

> DN<sub>st</sub> = Enhanced DN value <br /> 
> N<sub>j</sub> = Number of pixels having DN values in the j<sup>th</sup> range, in the input image <br /> 
> K = Maximum number of DN ranges in the input image <br /> 
> N = Total number of pixels in the input image

### Links: 
- https://en.wikipedia.org/wiki/Histogram_equalization
- https://theailearner.com/2019/04/01/histogram-equalization/

---
# Code for Equalized histogram contrast enhancement

```r
# Title    :: Compute Histogram from local image
#  Author  :: Ankit Deshmukh
#  DOC     :: 2022-08-30 10:46:37
#  DOLE    :: 2022-08-30

# https://rdrr.io/bioc/EBImage/man/equalize.html
# install.packages("BiocManager")
# BiocManager::install("EBImage")
library(EBImage)

# Sample - 01
x = readImage(system.file('images', 'cells.tif', package = 'EBImage'))
hist(x)
y = equalize(x)
hist(y)
display(y, title = 'Equalized Grayscale Image')
```

---
## Cont...

```r
# Sample - 02
x = readImage(system.file('images', 'sample-color.png', package = 'EBImage'))
hist(x)
y = equalize(x)
hist(y)
display(y, title = 'Equalized Grayscale Image')

par(mfrow = c(1,2))
plot(x)
plot(y)

par(mfrow = c(1,2))
hist(x)
hist(y)
```

---
## Cont...

```r
# Sample - 03: Import local images
x = readImage('./images/image-1.jpg')
hist(x)
y = equalize(x)
hist(y)
display(y, title = 'Equalized Grayscale Image')

par(mfrow = c(1,2))
plot(x)
plot(y)
```

---
# Refrence  
- https://rdrr.io/bioc/EBImage/man/equalize.html