A Data Augmented Method for Plant Disease Leaf Image Recognition based on Enhanced GAN Model Network

Xin, Ming Yuan; Ang, Ling Weay; Palaniappan, Sellappan

doi:10.33093/jiwe.2023.2.1.1

Cited by 8 publications

(3 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Plant disease detection is a critical domain to be investigated as they contribute to sustainable agriculture, higher crop yields, lower costs, and better resource management [23]. Through this research, we have addressed the critical challenge of higher accuracy with lower computational cost in a small imbalanced dataset with complicated backgrounds.…”

Section: Discussionmentioning

confidence: 99%

Plant Disease Detection and Classification Using Deep Learning Methods: A Comparison Study

Chin,

Ng,

Palanichamy

2024

JIWE

View full text Add to dashboard Cite

The presence issue of inaccurate plant disease detection persists under real field conditions and most deep learning (DL) techniques still struggle to achieve real-time performance. Hence, challenges in choosing a suitable deep-learning technique to tackle the problem should be addressed. Plant diseases have a detrimental effect on agricultural yield, hence early detection is crucial to prevent food insecurity. To identify and categorise the indications of plant diseases, numerous developed or modified DL architectures are utilised. This paper aims to observe the performance of the YOLOv8 model, which has better performance than its predecessors, on a small-scale plant disease dataset. This paper also aims to improve the accuracy and efficiency of plant disease detection and classification methods by proposing an optimised and lightweight YOLOv8 architecture model. It trains the YOLOv8 model on a public dataset and optimises the YOLOv8 algorithm with the integration of the GhostNet module into the backbone architecture to cut down the number of parameters for a faster computational algorithm. In addition, the architecture incorporates a Coordinate Attention (CA) mechanism module, which further enhances the accuracy of the proposed algorithm. Our results demonstrate that the combination of YOLOv8s with CA mechanism and transfer learning obtained the best result, yielding score of 72.2% which surpassed the studies that utilised the same dataset. Without transfer learning, our best result is demonstrated by YOLOv8s with GhostNet and CA mechanism yielding a score of 69.3%.

show abstract

Section: Discussionmentioning

confidence: 99%

Plant Disease Detection and Classification Using Deep Learning Methods: A Comparison Study

Chin,

Ng,

Palanichamy

2024

JIWE

View full text Add to dashboard Cite

show abstract

“…The smoothed signals are then fed into a deep learning network, which is trained to perform gender estimation based on the gait features extracted from the landmark positions. By utilizing deep learning algorithms [1], the proposed method effectively captures complex patterns and relationships within the gait features, resulting in enhanced accuracy and reliability of gender recognition.…”

Section: Introductionmentioning

confidence: 99%

GenReGait: Gender Recognition using Gait Features

Ti,

Connie,

Goh

2023

JIWE

View full text Add to dashboard Cite

Gender recognition based on gait features has gained significant interest due to its wide range of applications in various fields. This paper proposes GenReGait, a robust method for gender recognition utilizing gait features. Gait, the unique walking pattern of individuals, contains distinct gender-specific characteristics, such as stride length, step frequency, and body posture, making it a promising modality for gender estimation. The proposed GenReGait method begins by extracting landmark positions on the human body using a human keypoint estimation technique. These landmarks serve as informative cues for estimating gender based on their spatial and temporal characteristics. However, environmental factors can impact gait patterns and introduce fluctuations in landmark points, affecting the accuracy of gender estimation. To overcome this challenge, GenReGait introduces a robust preprocessing technique known as Weighted Exponential Moving Average to smoothen the gait signals and reduce noise caused by environmental factors. The smoothed signals are then fed into a deep learning network trained to perform gender estimation based on the gait features extracted from the landmark positions. By leveraging deep learning algorithms, the proposed GenReGait method effectively captures complex patterns and relationships within the gait features, enhancing the accuracy and reliability of gender recognition. Experimental evaluations conducted on the Gait in the Wild dataset and a self-collected dataset validate the robustness and effectiveness of the proposed GenReGait approach.

show abstract

“…Modern agriculture faces challenges like excessive chemicals to be used for high yield and pest control, which necessitates prompt and accurate disease diagnosis [1], [2]. Due to the industry's growth, traditional methods of diagnosing diseases based on farming expertise or professional advice are no longer sufficient.…”

Section: Introductionmentioning

confidence: 99%

Processing Plant Diseases Using Transformer Model

Marcus Lye,

2023

JOIV : Int. J. Inform. Visualization

View full text Add to dashboard Cite

Agriculture faces challenges in achieving high-yield production while minimizing the use of chemicals. The excessive use of chemicals in agriculture poses many problems. Accurate disease diagnosis is crucial for effective plant disease detection and treatment. Automatic identification of plant diseases using computer vision techniques offers new and efficient approaches compared to traditional methods. Transformers, a type of deep learning model, have shown great promise in computer vision, but as the technology is still new, many vision transformer models struggle to identify diseases by examining the entire leaf. This paper aims to utilize the vision transformer model in analyzing and identifying common diseases that hinder the growth and development of plants through the plant leave images. Besides, it aims to improve the model's stability by focusing more on the entire leaf than individual parts and generalizing better results on leaves not in the image center. Added features such as Shift Patch Tokenization, Locality Self Attention, and Positional Encoding help focus on the whole leaf. The final test accuracy obtained is 89.58%, with relatively slight variances in precision, accuracy, and F1 score across classes, as well as satisfactory model robustness towards changes in leaf orientation and position within the image. The model's effectiveness shows the vision transformer's potential for automated plant disease diagnosis, which can help farmers take timely measures to prevent losses and ensure food security.

show abstract

A Data Augmented Method for Plant Disease Leaf Image Recognition based on Enhanced GAN Model Network

Cited by 8 publications

References 18 publications

Plant Disease Detection and Classification Using Deep Learning Methods: A Comparison Study

Plant Disease Detection and Classification Using Deep Learning Methods: A Comparison Study

GenReGait: Gender Recognition using Gait Features

Processing Plant Diseases Using Transformer Model

Contact Info

Product

Resources

About