View of Weed Detection Using Image Processing

(1)

16130 http://annalsofrscb.ro

Weed Detection Using Image Processing

M Anand^1*, P Vijaya Mary², G Archana³, P Charith⁴

1Assistant Professor, Geethanjali College of Engineering and Technology, Hyderabad India

2, 3, 4 Students, Geethanjali College of Engineering and Technology, Hyderabad India

ABSTRACT

As the increase in the world population the demand of the Soya production and other is also increases. In order to increase the growth of the Soya in the Soya crop it is necessary to detect the weed in the Soya crop and the barren land to minimize the growth of weed so that the growth of the Soya can be improved. Weed detection is one of the important factors to be analysed. Unmanned Air Vehicle (UAV) is used to get data acquisition of Soya crop in different phases so that high quality of RGB images can be captured. The proposed method facilitates the extraction of weed, Soya, and barren land in the Soya crop field using background subtraction. The result shows that background subtraction method is good for detection the weed, barren land, and Soya.

Key words: UAV, RGB, CNN

INTRODUCTION

Weeds are an all too common occurrence in lawns and gardens. While some may be deemed useful or attractive, most types of weeds are considered a nuisance. Learning more about weed control and detection can make it easier for gardeners to decide whether these weeds should be welcomed or if they must go. Let’s take a look at some common weed plants and when or what weed control methods may be necessary. By definition, a weed is known as “a plant in the wrong place.” For the most part, these plants are known more for their undesirable qualities rather than for their good ones, should there be any. Weeds are competitive, fighting your garden plants or lawn grass for water, light, nutrients and space. Most are quick growers and will take over many of the areas in which you find them. While most types of weeds thrive in favourable conditions, native types may be found growing nearly anywhere the ground has been disturbed. In fact, they may even offer clues to your current soil conditions.

One of the newest and most researched technologies nowadays is deep learning. Deep learning is a technique used to create intelligent systems as similar as possible to human brains. It has made

(2)

a big impact in all types of domains such as video, audio and image processing. On the other hand, agriculture is humanity’s oldest and most essential activity for survival. The growth of population during the last years has led to a higher demand of agricultural products. To meet this demand without draining the environmental resources the agriculture uses, automation is being introduced into this field. The present project aims to merge both concepts by achieving autonomous weed recognition in agriculture this goal will be reached by using new technologies such as Open CV, FarmBot and Python programming, image processing, deep learning and Artificial Neural Networks (ANNs).

LITERATURE REVIEW

Agriculture has always been an essential activity for survival. Over the last century, and more specific, over the last 15 years, agriculture has started to mechanise and digitise due to this evolution and automation, labour flow was almost totally standardised. Nowadays, after introducing robotics and artificial intelligence into agriculture there is no need of standardization, robots are working collaboratively with humans and learning from them how to realize the basic agriculture tasks such as weed detection, watering or seeding.

Weed detection is one of those basic agriculture tasks that are being automatized and digitised, in this case, because of toxicity related to herbicides so, reducing human intervention will make possible a decrease in the use of herbicides, increasing health care. To achieve this, robots able to detect plants and classify them into crop or weed are now introduced into agriculture. This implementation has been done in multiples studies such as Dankhara. Where Internet of Things (IoT) is applied into an intelligent robot to differentiate crop and weed remotely IoT is present in the communication between a Raspberry Pi, where the processing is done and the camera and sensors are connected, and the Data Server, where the Raspberry Pi sends the information obtained.

Machine learning (ML)

Machine learning is the scientific study of algorithms and statistical models that computer systems use to perform a specific task without using explicit instructions, relying on patterns and inference instead. It is seen as a subset of artificial intelligence. Machine learning algorithms build a mathematical model based on sample data, known as "training data", in order to make

(3)

predictions or decisions without being explicitly programmed to perform the task. Machine learning algorithms are used in a wide variety of applications, such as email filtering and computer vision, where it is difficult or infeasible to develop a conventional algorithm for effectively performing the task.

CNN ALGORITHM IMPLEMENTATION

Convolutional Neural Network has had ground breaking results over the past decade in a variety of fields related to pattern recognition; from image processing to voice recognition. The most beneficial aspect of CNNs is reducing the number of parameters in ANN. This achievement has prompted both researchers and developers to approach larger models in order to solve complex tasks, which was not possible with classic ANNs. The most important assumption about problems that are solved by CNN should not have features which are spatially dependent. In other words, for example, in a face detection application, we do not need to pay attention to where the faces are located in the images. The only concern is to detect them regardless of their position in the given images. Another important aspect of CNN is to obtain abstract features when input propagates toward the deeper layers.

SUPPORT VECTOR MACHINE ALGORITHM

Support Vector Machine or SVM is one of the most popular Supervised Learning algorithms, which is used for Classification as well as Regression problems. However, primarily, it is used for Classification problems in Machine Learning. The goal of the SVM algorithm is to create the best line or decision boundary that can segregate n-dimensional space into classes so that we can easily put the new data point in the correct category in the future. This best decision boundary is called a hyper plane. SVM chooses the extreme points/vectors that help in creating the hyper plane. These extreme cases are called as support vectors, and hence algorithm is termed as Support Vector Machine. Consider the below diagram in which there are two different categories that are classified using a decision boundary or hyper plane.

METHODOLOGY

This project work describes the methodology for weed detection process in soya crops. It follows ten steps. The task flow is shown in Figure 1. The tasks and process represents the steps of the algorithm. Our algorithm includes the following steps

(4)

Fig: 1 Flow chart

(5)

Method of Data Collection

In this project, we attempt to build a classifier that can identify the weed based on the module we built. The data collection is the primary data which we take data manually by our self. The data collected has been check one by one before taking it as input data. There are 2 classes named as crop and crop weed will be classified and each of the class contains 123 images. From the 80%

of images used for training and 20% of images used for validation and testing. Hence there are 99 images for training and 24 images for testing

.

RESULTS

Fig: 2 Soya crop detection.

(6)

Fig: 3 Wed detection in soya crop.

Fig: 4 Percentage of soya and weed detection.

(7)

CONCLUSION AND FUTURE WORK

Although the detection was good enough but still there exist a room for the improvement in the result. Convolution Neural Network (CNN) can be applied in the future research for better result.

In stage when the weed and Soya colour are same convolution network (CNN) can be produced much better result. The limitation in the above proposed research that was faced was sunlight light intensity issue because of the variation in the captured image was quite high. This can be overcome when the similar research was done in a very controlled environment. Better camera can be used in order to capture much better image with different sensor in order to overcome the problem that arise due to sunlight variation and shadows.

REFERENCES

[1] K.Kantipudi , C.Lai , C.Hong. Min , Ron C. Chiang (2018). Weed Detection among Crops by Convolutional Neural Networks with Sliding Windows. Conference Paper.

[2] Ana I. de Castro, Jorge Torres-Sanchez, Jose M. Pena, Francisco M.Jimenez-Benes, Ovidu Csillik and Francisca Lopez-Granados(2018). An Automatic Random Forest-OBIA Algorithnm for Early Weed Mapping between and within Crop Rows Using UAV Imagery,21(17),pp.25-29.

[3] S. J. Leghari, U. A. Leghari, G. M Laghari, M. Buriro, F. A. Soomro, “An overview on various weed control practises affecting crop yield”, Journal of Chemical, Biological and Physical Sciences, Vol.6, No. 1, pp. 59-69, 2015

[4] International Survey of Herbicide Resistant Weeds, available at:

http://www.weedscience.org/default.aspx

[5] F. Ahmed, H. A. Al-Mamun, A. S. M. Hossain Bari, E. Hossain, P. Kwan, “Classification of crops and weeds from digital images: A support vector machine approach”, Crop Protection, Vol. 40, pp. 98-104, 2012

[6] S. Mallick, Blob Detection Using OpenCV (Python, C++), available at:

https://www.learnopencv.com/blob-detection-using-opencv-python

[7] M. Weis, M. Sokefeld, “Precision crop protection-the challenge and use of heterogeneity”, In:

Detection and Identification of Weeds, first ed. SpringerVerlag, pp. 119–134, 2010

[8] T. Rumpf, C. Romer, M. Weis, M. Sokefeld, R. Gerhards, L. Plumer, “Sequential support vector machine classification for small-grain weed species discrimination with special regard to

(8)

Cirsium arvense and Galium aparine”, Computers and Electronics in Agriculture, Vol. 80, pp.

89-96, 20

[9] P.Sakthi, P.Yuvarani(2018),Detection and Removal of Weed between Crops in Agricultural Field using Image Processing, 5(1), pp. 1-13.

[10] P.Sakthi, P.Yuvarani (2018). Detection and Removal of Weed between Crops in Agricultural Field using Image Processing.Electronics and Instrumentation Engineering, 118(8), 201-206.