15 research outputs found
Automated underwater vision system for detection and classification of marine life using CNN YOLO-based model / Mohamed Syazwan Asyraf Rosli
No full textThe exploration of underwater ecosystems requires thorough observation and investigation for different marine habitats leads to the usage of an underwater vision system. This method is superior as a non-destructive approach for video monitoring technique that takes advantage from the advancement of camera vision technology. However, manual processing of real-time or recorded video is a time-consuming, laborious, and highly possible for fatigue errors during analysis. This is due to the water murkiness and low-light projection that yield low visibility, hence limits the ability to precisely explore the underwater environments. Driven by these challenges, there is a need for an automated real-time method that can replace the manual analyzing process and be able to work under constraints environments. Recently, the integration of computer vision and machine learning has given solutions to improve the underwater detection system by using intelligent classifier algorithm in real-time computer vision to detect underwater animals with challenging environments. Therefore, this study is proposed to develop an automated underwater object detection model based on Deep Convolutional Neural Network (DCNN) by using You Only Look Once (YOLO). In this research, several single-stage detectors of YOLO models namely as YOLOv3, YOLOv4, YOLOv5 including their subset models were evaluated for benchmarking and comparison purpose. All models were trained and tested using three open-source datasets for assessing the models’ performances based on quantitative metrices and image processing speed. Hence, the proposed YOLO model is further improved based on the model optimization using a challenging The Brackish Dataset. The Adaptive Moment Estimation (Adam) optimizer and Learning Rate on Plateau are employed to optimize the model's training regime. Significantly, the comparison results show that YOLOv5s outperformed others in terms of mean average precision (mAP) up to 97.7% and inference speed of up to 125 Frame Per Second (FPS). Meanwhile, the Adam optimizer with a detailed learning rate and momentum fine-tuning provides sufficient convergence rate and assisting YOLOv5s achieve better performance of mAP which is 0.6% higher than SGD implementation and 13.72% better than original author for The Brackish Dataset implementation. As a conclusion, the proposed YOLO model is successfully improved with high precision performance for detecting underwater object in a challenging underwater environment. Thus, the model also provides sufficient image processing and speed capability at real-time which will help marine biologists in elevating underwater research for analyzing underwater video with minimal human intervention
The concept of tarikah in the Islamic law of succession with special reference to the practices of the civil courts and the Syariah courts in Malaysia
No full textEThOS - Electronic Theses Online ServiceGBUnited Kingdo
Enhancing Ai-driven personalized exercise and nutrition planning for elderly through prompt refinement and expert consensus / Nur Athirah Rosli ... [et al.].
No full textThis study aims to refine the NExGEN Prompt Generator–ChatGPT Framework for personalized exercise and nutrition planning tailored to Malaysia's elderly population using the Fuzzy Delphi Method [1]. Addressing gaps in AI-driven health interventions, the research focuses on enhancing prompt accuracy, scalability, and adaptability to meet elderly-specific health needs effectively. A convenience sample of 18 elderly (>60 yrs old) Malaysians. A purposive sample of 21 experts was recruited to evaluate the NExGEN framework using a custom-designed questionnaire based on personalized nutrition and exercise constructs. The Fuzzy Delphi Method was employed for consensus-building, with responses analyzed using Triangular Fuzzy Numbers and defuzzification techniques to assess expert agreement [2]. Expert feedback, collected through Likert scales and open-ended responses, informed iterative framework improvements
Effect of glufosinate-ammonium, glycophosphate and imazapyr herbicides at two spraying volumes on Imperata cylindrica (L.) Raeuschel
No full textImperata cylindrica (L.) Raeuschel is a problematic weed of the tropical region which causes significant losses in cultivated and non-cultivated areas, and is difficult to control. A field study was conducted to evaluate the effectiveness of glufosinate-ammonium, glyphosate and imazapyr herbicides on a stand of the weed in the field. The herbicide treatments were applied at their recommended field rates (glufosinate-ammonium at 3 kg a.i. ha-1; glyphosate at 1.65 kg a.e. ha-1; imazapyr at 0.75 kg a.e. ha-1), at two spray volumes of 400 and 800 L ha-1. Visual observation on foliage kill was related to the tiller number/unit area and mass production of the green foliage and fresh rhizomes, and the short-term plant succession pattern was also evaluated. Spray volumes of 400 and 800 L ha-1 did not influence the effectiveness of the herbicides on the weeds, however, significant kill (>80% kill) of the foliage was observed for each herbicide when compared with the control. Glufosinate-ammonium caused fast kill (within 1 week after treatment (wat)) of I. cylindrica foliage, but plant regeneration was faster (within 10 wat) than treatments with glyphosate or imazapyr which killed both the foliage and rhizomes (within 3 and 13 wat, respectively), resulting in longer control period and minimal regrowth of the weed. Regeneration of other plant species was much less in imazapyr treatment compared with those of the glyphosate or glufosinate-ammonium. Plants regenerated were mostly of the broadleaf species
Weed Control Efficacy and Crop-Weed Selectivity of a New Bioherbicide WeedLock
No full textPlant-based bioherbicides could be an effective alternative to current chemical herbicides for sustainable agriculture. Therefore, this research evaluated the weed control efficacy and crop-weed selectivity of the new plant-derived bioherbicide WeedLock compared to commercial herbicides in glasshouse and field conditions. In the glasshouse, the herbicides applied were WeedLock (672.75, 1345.50, 2691.00 L ha−1), glyphosate isopropyl-amine, and glufosinate-ammonium (1, 2, 4 L ha−1), over the untreated (control) on six weeds and four crops. In the field trial, typical weeds were allowed to grow at a uniform density across plots (2.5 × 2.5 m), and WeedLock (1345.50, 2691.00 L ha−1), glyphosate isopropyl-amine, and glufosinate-ammonium (2, 4 L ha−1) were applied along with untreated plot (control). A randomized complete block design was set with four replications for glasshouse and field experiments. WeedLock at 1345.50 L ha−1 showed efficacy similar to glyphosate isopropyl-amine and glufosinate-ammonium at 2 L ha−1 for Ageratum conyzoides L. in the glasshouse. Applied herbicides killed all tested crops except Zea mays L. at 1345.50 L ha−1 (WeedLock). WeedLock showed more than 50% efficacy at 35 days after spray, while 65% was produced by glyphosate isopropyl-amine and glufosinate-ammonium compared to the untreated plot (control). WeedLock has excellent potential to control weeds in both glasshouse and field conditions and showed a non-selective character
Design Improvements of Acoustic Response of an Automotive Air Intake System
No full textThis study aims to improve the acoustic response of an automotive air intake system (AIS) using harmonics acoustics analysis system. Noise pollution imposes adverse health effects on human beings. One of the sources of noise pollution comes from the transportation sector. Combustion engines produce noise through the act of combustion during operation. Noise generated by the AIS affects the comfort and long-term health of humans during operations. The noise produced by the AIS can be reduced through the introduction to resonators. A baseline result was modelled through creating an internal cavity and running a harmonics acoustic analysis simulation in ANSYS. Transmission loss and frequency response data was analysed as well as the location of the noise through visual contour. Modifications to the baseline geometry were done through the addition of a resonator targeted at 197 Hz and the geometry was then analysed again through the same parameters with harmonics acoustics analysis. The results show the resonator is effective with the peak in noise at 197 Hz reduced but additional noise is introduced as a result of the addition of a resonator into the AIS
Bioherbicides: An Eco-Friendly Tool for Sustainable Weed Management
No full textWeed management is an arduous undertaking in crop production. Integrated weed management, inclusive of the application of bioherbicides, is an emerging weed control strategy toward sustainable agriculture. In general, bioherbicides are derived either from plants containing phytotoxic allelochemicals or certain disease-carrying microbes that can suppress weed populations. While bioherbicides have exhibited great promise in deterring weed seed germination and growth, only a few in vitro studies have been conducted on the physiological responses they evoke in weeds. This review discusses bioherbicide products that are currently available on the market, bioherbicide impact on weed physiology, and potential factors influencing bioherbicide efficacy. A new promising bioherbicide product is introduced at the end of this paper. When absorbed, phytotoxic plant extracts or metabolites disrupt cell membrane integrity and important biochemical processes in weeds. The phytotoxic impact on weed growth is reflected in low levels of root cell division, nutrient absorption, and growth hormone and pigment synthesis, as well as in the development of reactive oxygen species (ROS), stress-related hormones, and abnormal antioxidant activity. The inconsistency of bioherbicide efficacy is a primary factor restricting their widespread use, which is influenced by factors such as bioactive compound content, weed control spectrum, formulation, and application method
Mokhtar
No full textMokhtar regains his consciousness in the forest. He attempts to get up but is unable to do so. He
hauls himself up to the nearest body for supplies but does not find much off of his dead comrade except for a rifle. Movement in the undergrowth catches his eye and upon realising it was an enemy soldier.
(Hiroo), Mokhtar attempts to attack him. Mokhtar fails to injure Hiroo and after realising that he is a medic, he demands that Hiroo hands over his supplies but the medic has other ideas and attempts to kill him instead. A short tussle ensues with the medic being knocked out cold and Mokhtar fatigued. When they both regain consciousness, they realise they would need to rely on each other to survive. As they interact with each other, they begin to gain each other’s trust and eventually, Hiroo patches up Mokhtar's leg wound.Bachelor of Fine Arts in Media Ar
Development of automated exhaust fan for modern kitchen with IoT notification system
No full textThis project shows the development of an automated exhaust fan for a modern kitchen with an IoT notification system for a better air ventilation system with lower electrical consumption and cheaper installation costs. Utilizing an exhaust fan as the ventilation system will significantly lower the cost of installation and maintenance. The exhaust fan will carry the process of removing hot air from the kitchen so that cooler surrounding temperatures can be achieved. The application of an automated microcontroller system will allow higher energy efficiency for daily cooking appliance usage. Then, the application of the DHT22 temperature sensor as the input for the microcontroller which can detect the precise temperature of the surrounding air. The exhaust fan will automatically turn ON when it detects the surrounding temperature is increased from 33℃. However, a notification will be sent to the user through the Blynk Application when the temperature of the kitchen is increasing and achieved 37℃ while a buzzer will turn ON to alert people in the house. Next, the AC voltage controller will help regulate the speed of the fan according to the input temperature so that more energy can be saved during long daily usage. At the end of this project, the outcome product will become a microcontroller that can be used in various types of households that already have a wall-mounted exhaust fan. Moreover, more users will be able to install this whole ventilation system for their hot kitchen due to low installation and maintenance costs. Lastly, the electrical monthly bill can be reduced
