DSpace Collection:
https://elibrary.khec.edu.np:8080/handle/123456789/405
2024-02-23T09:03:02ZRECOGNITION AND SEPARATION OF FRESH AND ROTTEN FRUITS USING YOLO ALGORITHM
https://elibrary.khec.edu.np:8080/handle/123456789/666
Title: RECOGNITION AND SEPARATION OF FRESH AND ROTTEN FRUITS USING YOLO ALGORITHM
Authors: Chet Narayan Mandal (750408); Kushal Shrestha (750412); Prem Bahadur Rana (750422); Ujjwal Dahal (750429)
Abstract: Fruit quality evaluation is crucial in today's food processing and distribution systems to assure consumer safety and minimize food waste. The fruits are often sorted manually using visual examination, which is time-consuming, labor-intensive, and prone to error. To address these issues, we can use the capabilities of computer vision and machine learning to build a powerful and real-time fruit quality evaluation system. This project presents an innovative approach for the automated detection and classification of fresh and rotten fruits through object detection. This project demonstrates a novel approach for the automated detection of fresh and rotten fruits on conveyor belts using the YOLO algorithm and Raspberry Pi.
The YOLOv7 object detection model was used for deep learning. The trained ‘Model’ is capable of recognizing the color and texture of the fruit surface as features to say whether it is rotten or fresh. The map of our trained model is 0.986%, F1_score is 0.96, precision is 0.957 And recall is 0.966. The main focus of this project is automation of fruit sorting through conveyor belt by detecting and recognizing rotten and fresh fruits.2023-08-01T00:00:00ZREAL-TIME ENCRYPTION AND DECRYPTION USING FPGA
https://elibrary.khec.edu.np:8080/handle/123456789/665
Title: REAL-TIME ENCRYPTION AND DECRYPTION USING FPGA
Authors: Ajay Shrestha (750404); Deepa Rajbhandari (750409); Rijan Shrestha (750424); Simran Giri (750427)
Abstract: This project “REAL-TIME ENCRYPTION AND DECRYPTION USING FPGA”
presents a novel approach to real-time encryption and decryption using the PYNQ FPGA
board in conjunction with the ESP32 WROOM module. Our project focuses on achieving
secure and efficient data communication, leveraging the capabilities of both the PYNQ
FPGA board and the ESP32 WROOM. We implement AES encryption algorithms to ensure
data privacy during transmission, and we demonstrate real-time performance by integrating
hardware acceleration on the PYNQ FPGA board. By offloading the cryptographic
operations onto dedicated hardware blocks within the FPGA, the system achieves reduced
latency and minimized processing overhead on the wireless communication module. We
applied optimization techniques, including pipelining and parallelization, to maximize the
throughput of the encryption process while adhering to the resource constraints of the FPGA.
Additionally, we explore the integration of the AES-enabled FPGA module into wireless
communication systems, highlighting its compatibility with various wireless communication
protocols. The combination of these technologies offers a powerful solution for secure realtime
data exchange, with potential applications in IoT, wireless communication, and other
domains requiring secure data transfer.
In order to compare our FPGA-accelerated technique with software-based AES
implementations, we analyzed latency, throughput, and resource usage. The outcomes
highlight the potential of FPGA-based AES encryption in boosting the security and
effectiveness of wireless communication by demonstrating significant improvements in
latency and data rate. We demonstrate the effectiveness and efficiency of our real-time
encryption and decryption system through experimental validation, demonstrating its
potential to improve data security.2023-08-01T00:00:00ZMODULAR GROUND STATION FOR AUTOMATIC TRACKING AND COMMUNICATION WITH LEO SATELLITE
https://elibrary.khec.edu.np:8080/handle/123456789/664
Title: MODULAR GROUND STATION FOR AUTOMATIC TRACKING AND COMMUNICATION WITH LEO SATELLITE
Authors: Abiral B.C (750431); Sajan Duwal (750425); Uran Shrestha (750430)
Abstract: The growing interest in satellite-based services requires a reliable and efficient communication link between ground stations and low-earth orbiting satellites. A CubeSat in low Earth orbit (LEO) within altitudes about 350km to 700km exhibits shorter orbital periods compared to satellites at higher altitudes. This feature of the LEO satellite has significant benefits in space missions. However, the line-of-sight (LOS) between a ground station and LEO satellites is limited due to their rapid orbital motion across the sky as they orbit the Earth. Hence, a network of ground stations is needed for efficient communication with LEO Satellites.
This project “Modular Ground Station for Automatic Tracking and Communication with LEO Satellite” aims to enhance LEO satellite communication using Modular Ground Station, offering a versatile and cost-effective solution for reliable and efficient communication with satellites. The Modular Ground Station architecture is a scalable and modular design that integrates with existing ground station infrastructures. The satellite tracking software utilizes the Two-Line Element (TLE) of the interested satellite to predict and track satellite movements. The ground station includes antennas, Software-Defined Radio (SDR), a tracking mechanism, Raspberry Pi, and software modules for satellite data processing, and signal decoding ensuring reliable and high-quality communication with LEO satellites. The NOAA satellites usually have Automatic Picture Transmission (APT) downlink mode of frequency around 137 MHz. Developing a Very High Frequency (VHF) antenna, the signal from the NOAA satellite is received and post-processed to an image.2023-08-01T00:00:00ZIoT-BASED SMART FISH FARMING AND AUTOMATION USING LoRa
https://elibrary.khec.edu.np:8080/handle/123456789/663
Title: IoT-BASED SMART FISH FARMING AND AUTOMATION USING LoRa
Authors: NAYAN BISTA (750417); PHURBA SHERPA (750420); SANJAYA LAMA (750426); SUPRIM LAMICHHANE (750428)
Abstract: This project focuses on the development of the remote monitoring and automation system for fish farming through the utilization of LoRa technology. Also, our primary objective is to create the system capable of monitoring multiple fish farming ponds minimizing the use of the internet. Usually farming places are quite far away from the resident area and where the internet availability is low. Moreover, the people that work in the fish farming ponds have to be engaged in all day activities to maintain the living fish habitat. So, this project focuses in less use of internet, less involvement of the human manpower, increase in the production.
In our system, various sensors like temperature, turbidity to sense the data from the pond in the real time. Generally, oxygen is main parameter required for the fish to live and for their growth and morning time is the best time to provide the oxygen in the farming ponds RTC was used for this purpose to automate the air pump for providing oxygen. If the sensor data is above the threshold value, then it triggers the relay and hence the relay turns on aerator and water pump to maintain the specific parameter. And similarly, feeding is also done automated and use of ultrasonic sensor to measure the food available in the food dispenser.
The purpose of our project is to increase in producing high quality fish by maintaining water parameters suitable for the fish growth in the fish tank.2023-08-01T00:00:00Z