The various fields
and applications can be analyzed by using digital image processing. In
agriculture sector the quality and quantity are the important measures from the
farmers perspective. Soil is the one of most important valuable resource . Soil pH is used to measure the acidity
and alkalinity in soils. The soil pH value of 7.0 is referred as
neutral, soil above this value is called as alkaline and soil belove this value is called as acidic. Soil
samples were collected and their pH was determined by using PCA algorithm.
formatting; style; styling; insert (key words)
I. Introduction (Heading 1)
Soil is recognized as one of the
most valuable natural resource. Soils are considered as the integral part of
the landscape and their characteristics are largely governed by the landforms
on which they have developed .Systematic study of soils provides information on
nature and type of soils for various uses. The pH in soils is an important
concerning part of the soil health. pH is a term that is used to describe the degree
of acidity or basicity. Soil acidity or alkalinity directly affects plant
growth. If a soil is too sour or too sweet, plants cannot take up nutrients
like nitrogen (N), phosphorus (P) and potassium (K). Most nutrients that plants
need are readily available when the pH of the soil solution ranges from 6.0 to
7.5. Below a pH of 6.0 (acid): Some nutrients such as nitrogen, phosphorus, and
potassium are less available. Above a pH of 7.5 (very alkaline), Iron,
manganese, and phosphorus are less available. A
soil with a pH of 6.0 is ten times more acidic than a soil of pH 7.0. Changes in
soil pH dramatically
affect the availability of nutrients to growing
crops. The pH meter is the preferred method
for determination of soil pH. A soil
analysis is a process by which elements such as P, K, Ca, MG, Na, S, Mn, Cu, Zn
are chemically extracted from the soil and measured for there “plant available
” content within the soil sample. The soil pH reflect whether a soil is acidic,
basic or alkaline. The acidity, neutrality or alkalinity of a soil is measured
in terms of hydrogen ion activity of the soil water system .The negative
logarithm of the H ion activity is called pH and thus pH of a soil is a measure
of only the intensity of activity and not the amount of acid present. The pH
range normally found in soil varies from 3 to 9 .
Mathematically pH is represented
log 1/H= log H+
Following table shows soil pH and Interpretation
The soil pH can be determined from soil color using on
digital image processing techniques. in which digital photographs of the soil
samples were used for the analysis of soil pH. Soil color is visual perceptual
property corresponding in humans to the categories i.e. red, green, blue and
others. Soil colors are the parts of visual perceptual property where digital
values of red, green and blue (RGB) provide a clue for spectral signature
capture of different pH in soil.
Of Soil Ph And Nutrient Using Image Processing
Soil samples were collected and their pH was determined by using
digital image processing technique. Soil colour is visual perceptual property
corresponding in humans to the categories i.e red, green, blue and others. Soil
colours are the parts of visual perceptual property where digital values of
red, green and blue (RGB) provide a clue for spectral signature capture of
different pH in soil. For the capturing images, digital camera was used. On the
basis of RGB grey values, pixels properties and their digital correlations,
results showed that there was a clear cut gap in grey values of colours in the
images . Ranges of soil pH and pH index values were 7.30-7.50 and
0.0070-0.0261, respectively in deep brown colour. Similarly, soil pH range
varies from 6.80-7.04 and 5.58-6.58 in light yellowish and greenish colour
respectively while their corresponding pH index values were 0.0071-0.0451 and
0.0084-0.0239. Thus soil pH range varies from 7.30-7.50, 6.80-7.04 and
5.58-6.58 in deep brown colour, light yellowish colour and greenish colour
2. Soil Salinity Mapping and Monitoring in
Arid and Semi-Arid Regions Using Remote Sensing Technology
analysis was performed on three thousand and eight hundred soil sample data
from Thrissur district. Soil pH, Electrical conductivity, Organic Carbon, Phosphorus,
Potassium, Calcium, Magnesium, Sulfur, Zinc, Boron, Iron, Copper and Manganese
data were analyzed.
analysis, ANOVA and Principal Component analysis were performed on the data
set. Analysis indicate that different soil components are significantly
correlated with soil properties.
of Agriculture Soil by Digital Image Processing
soil pH is measured manually in Government Labs. There are various labs but
instrument used for this is not available everywhere. So we need to implement
calculation of soil pH by using digital Image Processing. Eighty soil samples
were collected and their pH firstly tested in Government Soil Testing Lab,
Agriculture College Nagpur and also it was determined by using digital image
processing technique. On the basis of RGB values, pixels properties and their
digital correlations, results showed that our pH values were approximately
matching with results from Government Testing lab.
Characterization Based on Digital Image Analysis
The digital image
database is prepared for the collected soil sample in the laboratory and
physical properties(Y) are determined. Digital image analysis is adopted to
estimate the image feature namely fractal dimension (X). Correlation is
developed between Y and X by fitting
appropriate polynomial equations using regression models.
The final results of this research will contribute to make soil physical
properties estimation automatic up to a certain degree of level, which will
assist to a geotechnical engineers, in soil classification. The results of
present work emphasizes that there is a great potential in the use of fractal
dimension for estimating physical soil properties for practical application, with
minimum human error.
processing scheme consists of image acquisition through digital camera or
scanner or mobile phone. Image processing includes image enhancement, filtering
of image to remove noise etc .image segmentation, feature extraction and
Detection. In the proposed work to design the system, training dataset and
test images are considered. The training dataset is the raw data which has
large amount of data stored in it and the test image is the input given for
involved in designing are as follow
In any of the image processing techniques the first
task is to acquire the image from the source. These images can be acquired
either through camera or through standard datasets that are available online.
The images should be in .jpg format. The images considered are user dependent.
is mainly done to eliminate the unwanted information from the image acquired
and fix some values for it, so that the value remains same throughout. In the
pre-processing phase the images are converted from RGB to Gray-scale and are
resized to 512*512 pixels. The images considered are in .jpg format, any other
formats found will not be considered for further processing. During
pre-processing, whole face is considered to be the region of interest. It is
detected by the cascade object detector which utilizes Jones-Viola algorithm.
Segmentation means partitioning of images into various
part or region and extracting meaningful region known as region of interest
(ROI).The level to which subdivision is carried depends on the problem being
solved .Segmentation can be stopped when the region of interest in an
been isolated. Segmentation accuracy determines success or
failure of computerized analysis procedures. So algorithm picked for
segmentation should perform best for given requirement.
After pre-processing the next
step is feature extraction. The extracted Soil features are stored as the
useful information in the form of vectors during training phase and testing
phase. There are various methods for extracting the Soil features some of which
are as listed below:
based Method: It is a rule based method that takes the knowledge of faces and
builds certain rules.
based Method: It extracts the features based on the sizes and positions of the
components from an image.
based Method: With the appropriate energy function, features are extracted from
previous template and the one that matches best yields minimum energy.
based Method: It takes skin color into account to bifurcate between the face
region and non face region from an image.
based Method: They represent important parts of the face because of their
holistic nature which deals with whole face. Here methods like PCA or ICA are
used to extract the Soil features.
Among all the above mentioned methods the one that is
used in the proposed work is PCA method for the reason that: it requires less
memory and storage capacity as the operation is carried out in a space of
smaller dimensions which also helps in increasing the efficiency of the system.
It is independent from the Soil geometry like size and shape of the face. The
possibility of real-time realization without even making use of any special
hardware’s like sensors or EEG equipment. It is easy to use and gives high
speed of recognition as compared to other methods. It gives higher success rate
in comparison with other methods.
A. PCA ALGORITHM
Principal component analysis (PCA) or karhunen-loeve transformation is
a statistical approach used for pattern recognition and signal
to reduce the number of variables in face recognition technique. PCA
technique has enormous potential as a feature extractor and is one of the
approaches to improve the reliability of the recognition systems. PCA is hugely
utilized in all forms of scrutiny because it is a simple method. It likewise
lessens the dimension of a figure i.e. from higher dimension to lower dimension
space effectively and yet holds the primary information of the images. In this technique,
every image in the training phase are represented as weighted eigenvectors that
are linearly combined and known as “Eigenfaces” and these eigenvectors are
obtained from the covariance grid of a training dataset which is known as basis
function. From the largely appropriate Eigenfaces the weights of an image are
obtained. Similarity between the pixels among images in a dataset by
means of their covariance matrix is one of the advantages taken by Eigen faces.
projecting the test image on the subspace spanned by the eigenfaces the
recognition of the Soil expression is performed and then the further
classification is carried out by a distance measure method known as Euclidean
Following are the steps involved to recognise the Soil
expressions using PCA approach:
1 Collection of Images to make the Database. Firstly the
images of various area of soil captured using digital camera with required
resolution for better quality or by scanner. The construction of database is
clearly dependent on application. All the databases are collected and stored in
2 Checking whether Image is colored or gray: Initially the
soil image is taken and is checked whether it is colored or gray image by using
the command size. If the soil image taken is colored then it is converted to
gray image to make it two dimensional (m x n). After this the class type of
gray image is checked to make sure the image type is doubled or not. If it is
not doubled than convert the image to double.
3 Mean and Standard Deviation of Images is Calculated: After
this the image is converted into a column vector of dimension mxn, so that a
typical image of size 112×92 for example becomes a vector of dimension 10304,
denoted as T. Now the mean of the image is calculated by using equation (4) and
standard deviation (S) is calculated using equation(13).
Also the standard deviation (ustd) and mean (um) is set
practically approximately close to mean and standard deviation calculated
4 Normalization: Next normalization of images is done
using equation (14) in order to make all images of uniform dimension.
5 Calculating Train Centred Images: Subtracting the mean
from column vector matrix of training images in order to obtain the centred
6 Calculating Eigen Vectors and Values from the covariance
7 Creating Eigen faces: Eigen Vectors obtained after SVD
are sorted in descending order and the top Eigen vectors are considered as
8 Calculating Train Weights: Now the weights are
determined by calculating the dot product of transpose of Eigen Faces matrix
and Train Centred Images.
9 Store Train Weights in Sink for Further Comparison: The
above mentioned procedure from step (2 to 8) is applied on each train image and
finally one by one train weights are stored in sink for further comparison.
10 Euclidean distance Classifier: In this block firstly all
steps from (2 to 7) are applied on the test image to compute its test weights
and then the difference between this test weights and the train weights stored
in the sink.
Recognized: Finally the minimum distance gives the best match or can be said
similar soil pH Identified.