Deep Learning-Based Mobile Application for Plant Disease Diagnosis

With the increasing computational power, areas such as machine learning, image processing, deep learning, etc. have been extensively applied in agriculture. This chapter investigates the applications of the said areas and various prediction models in plant pathology for accurate classification, identification, and quantification of plant diseases. The authors aim to automate the plant disease identification process. To accomplish this objective, CNN has been utilized for image classification. Research shows that deep learning architectures outperform other machine learning tools significantly. To this effect, the authors have implemented and trained five CNN models, namely Inception ResNet v2, VGG16, VGG19, ResNet50, and Xception, on PlantVillage dataset for tomato leaf images. The authors analyzed 18,160 tomato leaf images spread across 10 class labels. After comparing their performance measures, ResNet50 proved to be the most accurate prediction tool. It was employed to create a mobile application to classify and identify tomato plant diseases successfully.

Soft-Computing-Based Approaches for Plant Leaf Disease Detection

The disease occurrence phenomena in plants are season-based which is dependent on the presence of the pathogen, crops, environmental conditions, and varieties grown. Some plant varieties are particularly subject to outbreaks of diseases; on the other hand, some are opposite to them. Huge numbers of diseases are seen on the plant leaves and stems. Diseases management is a challenging task. Generally, diseases are seen on the leaves or stems of the plant. Image processing is the best way for the detection of plant leaf diseases. Different kinds of diseases occur because of the attack of bacteria, fungi, and viruses. The monitoring of leaf area is an important tool in studying physiological capabilities associated with plant boom. Plant disorder is usually an unusual growth or dysfunction of a plant. Sometimes diseases damage the leaves of plants.

Identification of Good One From the Damaged Crops/Fruits Using Decision-Level Information Matching

Identification of useful items that can be picked up from the damaged crops or batch of fruits/vegetables is a challenging task nowadays. Humans may fail to identify them correctly with their naked eyes due to strain. Image processing techniques can help to maximize the amount of the good agro-items easily by comparing the existing goods to templates. This chapter introduces an effective recognition method to spot good agro-items by extracting the local features using Gabor filter for orientation information. Another local information of that fruit/vegetable is extracted by speeded up robust features (SURF) algorithm. The extracted features are matched with their templates which results in the decision of individual feature extraction method. Finally, both local information is fused at decision level individually with AND operation (i.e., both algorithms will give correct decision to identify the good agro-item).

Intelligent Farming With Surveillance Agribot

This chapter endeavors to execute the rural cultivating robot which can move naturally, instinctually, automatically to furrow, seeding, and water system in shut field. This examination additionally adds to upgrade usefulness of agri-bot in the field for spying. The agri-robot is outfitted with a camera and sends the information to pc through wi-fi arrange. The robot has furrowing cutting edges and servo engine for dispersing the seeds and water system into the field. It works with ultrasonic sensor and IR sensor. Ultrasonic sensor is for dodging deterrents in the field; IR sensor is for detecting felid limit. Arduino controller goes about as heart and cerebrum of the framework; it makes quick, precise, self-governing development. This exploration includes an agri-bot that utilization Wi-Fi 802.11G alongside TCP/IP convention. IP address and video are gotten in workstation for further handling. This examination means to decrease human endeavors and give a clever guide to the farmers.

Application of Software in Soil and Groundwater Recharge Estimation in Ilorin, Nigeria

Rainfalls measured in a selected location in Ilorin, Nigeria and standard formula were used to fix the unknown parameters of the new numerical formula using Microsoft Excel Solver. The new numerical formula was used to estimate groundwater recharge from the rainfall. The accuracy of the new numerical formula was evaluated statistically and compared with the previous formulae in use using field groundwater recharge. Correlation between rainfall and estimated groundwater recharge was stablished. Annual cost benefit of groundwater recharge was computed. The study revealed that new formula provided the lowest relative error of 0.001%, the highest MSC of 17.747; the degree of accuracy of 99.999% correlation factor between rainfall and groundwater recharge using the new numerical formula was 0.1612 with correlation coefficient of 0.6079. The average annual cost benefit was1080.24 $ m-2 per year. It was concluded that modeling of groundwater recharge using the new numerical formula is a promising tool for estimating groundwater recharge with minimum error in water resources management.

Prognosis for Crop Yield Production by Data Mining Techniques in Agriculture

In agriculture, data mining technique is used for extracting information from a large dataset. The techniques for data mining are used in yield prediction for crop at broader spectrum. Agricultural system is very complex and vast therefore to deal with large data situation is a great factor. Different consultancy, industrial production department, organization related to crops is taking keen interest towards crop yield prediction. Here the focus is on the applicability of data mining techniques in agricultural field. The classification and clustering techniques of data mining are used recently in agriculture field. Data mining technology merged with the rapid development of computer science. This chapter focuses on collecting information and overcome the short comes of manual data handling and prediction of yield results of crop production. Data mining is a prominent agricultural research area for analysis of crop yield. These predictions are a very important in solving agricultural problems for crops.

Identification of Agricultural Crop Residues Using Non-Destructive Methods

Biomass is a bulky and inhomogeneous material, making it difficult to transport and store. In order to solve this problem, it has been found that the most common way to overcome the limitation of the biomass bulk density is to increase it with fine shredding. This chapter investigated the ability to identify specific operation conditions in a prototype biomass shredder by developing and utilizing non-destructive testing and artificial intelligence techniques. In order to demonstrate the performance of proposed methods, three different case studies investigated the different operation conditions from the vibration signals acquired through the ball bearings of the biomass shredder. The results showed that the two classifiers can provide reliable results using as inputs statistical features in time and frequency domain. These statistical features can be used with success for identify different operating condition. The combination of the statistical features with the appropriate classifiers gives a powerful tool for the agricultural biomass shredder condition monitoring.

WT-MO Algorithm

Visual examination of blood smears is an essential tool for analysis, prevention, and remediation of several types of maladies. The interest of computer-aided decision has been acknowledged in many medicinal instances (e.g., automatic ways and means are being explored to spot, classify, and measure visual items in hematological cytology [HC]). This chapter proposes an entirely automated blood smear diagnosis system for hemograms, which can lessen the time spent to scrutinize a slide. The present framework relies on morphological operations (MOs) and soft segmentation by means of the watershed transform (WT). Experiments demonstrate the method efficacy to count white blood cells (WBCs) and red blood cells (RBCs). Some considerations about implementations, design advice and possible variants, as well as improvements are discussed. The future of automated medical analysis is contemplated.

Applications of Image Compression on Agricultural Image Data Analysis

With the advent of the cutting-edge technologies in information repository and communication, data storage is rapidly increased, and it is required to reduce the size of the data. Especially in the case of agricultural image data like types of plants, crops, seeds; kinds of diseases and their remedial pesticides; and the agricultural satellite; images require a huge volume of memory space to store. To avoid this problem, it is required to reduce the size of the data and redundancy of the data. To overcome this problem, this chapter proposes a compression method, based on an adaptive Gaussian Markov random field model for agricultural image data compression, where the images are assumed to be a Gaussian Markov random field. The parameters of the model are estimated, based on the Bayesian approach. The authors use arithmetic coding to store seed values and parameters of the model as it augments the compression ratio. They also have studied the use of the M-H algorithm, which updates the parameters and through which the image contents such as untexturedness are captured.

Examination of Plant/Weed Image Dataset Using a Hybrid Image Processing Tool

Image evaluation procedures are widely employed in various domains to extract the useful information to make the necessary decision. This chapter implements a soft-computing tool to examine the benchmark plant/weed (BPW) pictures of computer vision problems in plant phenotyping (CVPPP2014) challenge database. The proposed work implements a hybrid image evaluation procedure based on social-group optimization algorithm (SGOA) and Shannon's entropy (SE)-based multi-thresholding and the Chan-Vese segmentation (CVS)-based extraction procedure. After extracting the crop/weed regions of BPW pictures, the superiority of the proposed tool is then assessed by implementing a comparative study between the extracted plant/weed region and its equivalent ground-truth. The results of this study substantiate that the proposed system is proficient in examining the BPW pictures and in future. This procedure can be considered to inspect the crop/weed pictures obtained with field supervising drones.