Convolutional neuralnetworks (CNNs) have been extensively studied for plant classification. Previous research primarily focused on herbs, plant diseases, and flowers. However, a limited amount of research specificall...
详细信息
ISBN:
(数字)9798350389692
ISBN:
(纸本)9798350389708
Convolutional neuralnetworks (CNNs) have been extensively studied for plant classification. Previous research primarily focused on herbs, plant diseases, and flowers. However, a limited amount of research specifically addresses the classification of ornamental plants based on their species. This study developed a classification system using a Residual Network with 50 layers (ResNet-50) implemented on a Raspberry Pi 3B. The system utilized a Raspberry Pi camera module for image capture and employed transfer learning with a pre-trained model from Keras. The model was trained in 6 classes, including Aglaonema commutatum, Dieffenbachia compacta, Spathiphyllum wallisii, Dracaena bacularis, Dracaena trifasciata, and unknown, using a dataset of over 200 images per class. Testing the model on 120 samples (20 per class) yielded an overall accuracy of 93.33%, as assessed by a confusion matrix. This paper is intended to aid the Bureau of Plant Industry for the benefit of research and development in the plant industries. This study also benefits plant retailers and buyers by ensuring accurate plant identification.
Genuine dialogue is encouraged in safe social environments. Conversations become more meaningful and real when people feel free to be themselves, voice their perspectives, and share their experiences without filtering...
详细信息
Object detection is a fundamental task in computer vision and image understanding, with the goal of identifying and localizing objects of interest within an image while assigning them corresponding class labels. Tradi...
详细信息
ISBN:
(数字)9798331506520
ISBN:
(纸本)9798331506537
Object detection is a fundamental task in computer vision and image understanding, with the goal of identifying and localizing objects of interest within an image while assigning them corresponding class labels. Traditional methods, which relied on handcrafted features and shallow models, struggled with complex visual data and showed limited performance. These methods combined low-level features with contextual information and lacked the ability to capture high-level semantics. Deep learning, especially Convolutional neuralnetworks (CNNs), addressed these limitations by automatically learning rich, hierarchical features directly from data. These features include both semantic and high-level representations essential for accurate object detection. This paper reviews object detection frameworks, starting with classical computer vision methods. We categorize object detection approaches into two groups: (1) classical computer vision techniques and (2) CNN-based detectors. We compare major CNN models, discussing their strengths and limitations. In conclusion, this review highlights the significant advancements in object detection through deep learning and identifies key areas for further research to improve performance.
Within the realm of sports competition, “momentum” may decisively influence the outcome of a game. Therefore, this study explores and examines “momentum” by utilizing real game datasets. First, this paper construc...
详细信息
ISBN:
(数字)9798350360240
ISBN:
(纸本)9798350384161
Within the realm of sports competition, “momentum” may decisively influence the outcome of a game. Therefore, this study explores and examines “momentum” by utilizing real game datasets. First, this paper constructs a model based on the cumulative score density during a match to capture the dynamics of the match. Second, in the calculation of cumulative momentum, the “inner area method” is innovatively adopted to mitigate the effects of unstable momentum fluctuations. Finally, gray relational analysis (GRA), wavelet analysis, and backpropagation neural network (BPNN) models are used to model and correlate the momentum prediction in tennis. The final experimental results show that the method proposed in this paper has strong validity and reliability.
image classification has widespread applications in various fields such as medical diagnosis, autonomous driving, security surveillance, and manufacturing. The existing architecture has a complex computation workload,...
详细信息
ISBN:
(数字)9798350390254
ISBN:
(纸本)9798350390261
image classification has widespread applications in various fields such as medical diagnosis, autonomous driving, security surveillance, and manufacturing. The existing architecture has a complex computation workload, leading to increased latency. This paper designs and implements a convolutional neural network architecture accelerated by FPGA, quantizing the parameters to int14 fixed-point integers, achieving an effective recognition rate of 98.2 % for handwritten digits. By reducing the number of parameters and arithmetic operations, the difficulty of deploying this CNN architecture on hardware devices is lowered, it also reduces the usage of storage resources, the combination of parallel computation of convolutions with pipelining operations improves the real-time performance of the overall architecture. Utilizing the resources on the ZYNQ-7020 FPGA development board to build the CNN architecture, achieving a prediction time of 86.02 μs for a single image under a clock frequency of 50 MHz, compared to existing architectures, it has significant advantages in terms of real-time performance.
This research investigates the limitations of current quantum hardware in fulfilling the computational demands of quantum neuralnetworks. It introduces an optimized circuit computed method leveraging bit splitting wi...
详细信息
ISBN:
(数字)9798350390643
ISBN:
(纸本)9798350390650
This research investigates the limitations of current quantum hardware in fulfilling the computational demands of quantum neuralnetworks. It introduces an optimized circuit computed method leveraging bit splitting within the framework of distributed quantum computing. The research exemplifies this approach by decomposing a 25-qubit variational quantum circuit into several smaller 5-qubit circuits, thereby enabling distributed quantum computations. A hybrid quantum-classical neural network architecture employing this methodology was implemented and evaluated on selected computer vision datasets. The results were promising, affirming the viability of the approach. The successful implementation of this architecture suggests that future distributed quantum computing systems will likely consist of scalable, heterogeneous quantum systems interconnected through an extensive quantum network. This configuration is anticipated to significantly enhance computational power and versatility, expanding the scope of quantum computing applications in real-world scenarios. This advancement not only validates the feasibility of distributed quantum architectures but also underscores their potential to revolutionize various technological domains through enhanced computational capabilities.
artificial intelligence has become more popular than ever in past few years. Deep neuralnetworks are being used in various use cases e.g., imageprocessing, data analysis and etc. MAC operations are the core of DNNs ...
详细信息
ISBN:
(纸本)9798350324532
artificial intelligence has become more popular than ever in past few years. Deep neuralnetworks are being used in various use cases e.g., imageprocessing, data analysis and etc. MAC operations are the core of DNNs thus making MAC units a very crucial element of any DNN accelerator. This paper presents a Variable precision, mixed fixed/floating point MAC unit capable of performing single precision floating-point MAC. Also, proposed MAC unit features additional modes for performing one 32-bit fixed-point MAC or two concurrent 16-bit fixed-point MACs or four concurrent 8-bit fixed-point MACs. Aside from high flexibility in number precision, proposed MAC unit uses recurring Karatsuba algorithm to implement higher bit-count multiplication only by using 8-bit multiplier and 8-bit adders. Proposed MAC unit has achieved 44.64 MOPS in 32-bit floating-point, 44.64 MOPS in 32-bit fixed-point, 89.29 MOPS in 16-bit fixed-point and 178.57 MOPS in 8-bit fixed-point on FPGA board `NEXYS 4 DDR' featuring XILINX 'xc7a100tcsg324-1' FPGA chip.
Deep learning is becoming more popular in practically every industry, but especially in medical imaging for better diagnostics of various deadly diseases. Deep learning is used to explain difficulties based on medical...
详细信息
3D human reconstruction from a single image has achieved great progress with recent deep neuralnetworks. However, conventional approaches still struggle with the issues of over-smoothing details and wrong limb poses....
3D human reconstruction from a single image has achieved great progress with recent deep neuralnetworks. However, conventional approaches still struggle with the issues of over-smoothing details and wrong limb poses. To this end, we propose PMDI, a method that combines parametric-model and depth-aware implicit function for single-view human reconstruction. First, given an RGB image of a person’s whole body as input, the method predicts its corresponding SMPL parameter model, depth map, and front (back) normal map by using deep neuralnetworks. Then, the predicted front depth map and normal feature are used as the additional parameters of the deep implicit function for reconstructing coarse results. Finally, the fine result is produced by integrating its corresponding coarse result with detailed back D-BiNI surface. Extensive experiments on the current large publicly available dataset (including DeepHuman and THUman2.0) demonstrate that PMDI outperforms the state-of-the-arts including PIFu, PIFuHD,PaMIR, and ICON.
The objective of the study is to automate the Structural Health Monitoring and site inspection work by developing and modeling an artificialneural Network with Convolutional neural Network technology using images of ...
详细信息
暂无评论