Arecanut Disease Detection Using CNN and SVM Algorithms

Sadanand

et al. 2023

IJEECS

Arecanut is one of the prominent commercial crops that are grown worldwide for traditional medicines, furniture, cosmetics, food, veterinary preparations, and textile industries. It experiences a variety of diseases during its existence, from the bottom to the tip. The conventional method for detection of diseases is through visual inspection and it is also necessary to have properly designed laboratories to check these harvests. It is a time consuming and tedious task to inspect these crops across wide acres of plantations. The proposed system has been developed that uses convolutional neural network (CNN) to identify and categorize diseases in arecanuts, trunks and leaves also suggesting effective preventative measures. Proprietary dataset consists of 1,100 photos of healthy and diseased arecas. The ratio between the train and test data is 80:20. Binary cross entropy is employed as the loss function for model construction, with accuracy serving as the metrics and Adam serving as the optimizing function. In identification and categorization of arecanut diseases, the suggested approach was shown to be efficient with 93.05% accuracy.

Section: Referencementioning

confidence: 99%

Identification and categorization of diseases in arecanut: a machine learning approach

Sadanand

et al. 2023

IJEECS

“…Excess nitrogen pollution from air pollution leads to eutrophication, which disrupts the balance of nutrients in water and negatively affects aquatic life. Acid rain, another consequence of air pollution, damages soil and further harms ecosystems [3][4] [5].…”

Section: Introductionmentioning

confidence: 99%

Iot Based Noise Detection and Management for Smart City

2023

IRJMETS

Air and noise pollution are escalating issues that demand continuous monitoring to ensure a healthy and safe environment. The rapid expansion of infrastructure and industrial plants has heightened the need for intelligent monitoring systems. The Internet of Things (IoT) has gained immense popularity due to its costeffectiveness, efficiency, and versatility, making it an ideal solution for addressing these challenges. IoT facilitates seamless communication between devices and humans, bridging the gap between human-machine interactions. Traditionally, data collection involved laborious field trips to various locations, followed by timeconsuming analysis. However, the integration of internet-connected sensors and microcontrollers has revolutionized environmental parameter monitoring, making it more flexible, accurate, and efficient. This paper explores the potential of IoT-based smart monitoring systems in effectively monitoring air and noise pollution, emphasizing the advantages of real-time data collection, remote accessibility, and automated analysis. The implementation of such systems can significantly contribute to efficient environmental management and enable timely interventions to mitigate pollution and safeguard public health.

“…Unsustainable fisheries are caused by illegal, unreported, and unregulated (IUU) fishing, which also threatens the health of marine ecosystems and has a negative impact on developing country economies [1] [2]. According to the Food and Agriculture Organization of the United Nations, approximately 91-93 million tonnes of fish are harvested worldwide each year, with roughly one-third of fisheries currently overfished and approximately 60 percent fished to capacity [3] [4]. IUU fishing is the world's sixth-largest crime, accounting for up to one-fifth of all fish caught at sea and estimated to be worth up to $23.5 billion per year [5] [6].…”

Section: Introductionmentioning

confidence: 99%

A Machine Learning Approach to Identifying and Preventing Illegal Fishing

2023

IRJMETS

Commercial fishing is the most common type of maritime human activity. Commercial fishing is the for-profit activity of harvesting fish or other marine species using boats, and its effects extend far beyond maintaining ocean health and managing fish populations. The monitoring of marine resources is a critical concern for marine resource management. Illegal fishing operations are prohibited. Overfishing and illegal, unreported, and unregulated (IUU) fishing pose serious threats to the sustainability of fish populations and other marine resources. Commercial fishing industry management has historically been complicated and costly, but technology and openness are bringing about real transformation. Using a variety of machine learning methods, one or more pursuers are trained to successfully catch an evader. This method is used to replicate actual illegal and unreported fishing chase events on both the pursuer and evader vessel agents. The goal of this effort is to reduce and eliminate cross fishing and illegal fishing. The Automatic Identification System, which is gathered by satellites and terrestrial receivers, may detect specific fishing activity based on the movement of vessels over time. Each vessel sends that information out on a regular basis.