Recently, various approaches using manually designed convolutional neural networks (CNNs) and U-shaped (encoder-decoder) models have yielded encouraging results in the automatic segmentation of medical images. Neverth...
详细信息
ISBN:
(纸本)9783031821554;9783031821561
Recently, various approaches using manually designed convolutional neural networks (CNNs) and U-shaped (encoder-decoder) models have yielded encouraging results in the automatic segmentation of medical images. Nevertheless, the manual construction of these models is laborious, requires a high level of expertise, is time-consuming, error-prone and can result in the loss of crucial details. To improve segmentation performance, these models often incorporate numerous parameters, which can lead to problems such as overfitting, vanishing and exploding gradient and increased computational complexity. In this study, we implement a genetic algorithm (GA) specifically designed for a U-Net architecture to segment images of skin lesions, using two datasets ISIC-SMALL1 and ISIC-SMALL2 constructed from ISIC-2017 dataset. The main objective is to optimize internal block topologies within the U-Net, in order to identify a high-performance deep neural network with a minimum of parameters. Compared to the U-Net baseline, our proposed methodology demonstrates favorable outcomes in terms of both the number of trainable parameters and Intersection over Union (IoU). For the ISIC-SMALL1 dataset, the leading model surpasses U-Net with a reduced parameter count (0.352001M) and an elevated IoU score (0.826052 vs. 0.736352). Similarly, the second-best model outperforms U-Net with fewer parameters (0.258781M vs. 31.043586M) and a higher IoU score (0.821956 vs. 0.736352). In the ISIC-SMALL2 dataset, both top models outshine U-Net in IoU (0.753122 and 0.722079 vs. 0.725363, respectively) while maintaining lower parameter counts.
The research presents a hybrid approach to identify and categorise nutritional deficiency syndrome in citrus leaves using imageprocessing and machine learning. The method includes processingimages, segmenting images...
详细信息
Based on human visual systems, imageprocessingalgorithms, and efficient hardware implementation methodologies are proposed to optimize the image qualities of AR displays according to the changes in ambient lights. T...
详细信息
ISBN:
(纸本)9798350327038
Based on human visual systems, imageprocessingalgorithms, and efficient hardware implementation methodologies are proposed to optimize the image qualities of AR displays according to the changes in ambient lights. To this end, methods are described to improve the image qualities perceived by humans. In addition, the delta look-up table is presented to minimize the number of additional circuits without significant changes in existing hardware. HOSA, an image quality assessment based on the human visual system is used to verify the image qualities for the extreme ambient light conditions.
Palm recognition systems play an important role in biometric authentication;however, existing systems frequently have low accuracy and resiliency due to problems such as changing lighting conditions, occlusions, and h...
详细信息
systems known as Automatic Number Plate Recognition (ANPR), license plate recognition or LPR are now widely used in many sectors such as law enforcement, traffic control, vehicle access etc. This is a technology that ...
详细信息
Addressing the limitations of currently rare small target detection algorithms based on Human Visual systems (HVS) that struggle with achieving satisfactory performance in complex backgrounds and lack high real-time c...
详细信息
Does progress on imageNet transfer to real-world datasets? We investigate this question by evaluating imageNet pre-trained models with varying accuracy (57% -83%) on six practical image classification datasets. In par...
详细信息
ISBN:
(纸本)9781713899921
Does progress on imageNet transfer to real-world datasets? We investigate this question by evaluating imageNet pre-trained models with varying accuracy (57% -83%) on six practical image classification datasets. In particular, we study datasets collected with the goal of solving real-world tasks (e.g., classifying images from camera traps or satellites), as opposed to web-scraped benchmarks collected for comparing models. On multiple datasets, models with higher imageNet accuracy do not consistently yield performance improvements. For certain tasks, interventions such as data augmentation improve performance even when architectures do not. We hope that future benchmarks will include more diverse datasets to encourage a more comprehensive approach to improving learning algorithms.
Accurate recognition of intra-pulse modulation patterns is essential for enhancing radar system performance. Tranditional recognition algorithms are typically designed under ideal conditions and handcrafted features, ...
详细信息
This work focuses on enhancing the quality of A- and B-scans of a novel linear optical coherence tomography system (LOCT), addressing the image degradation caused by noise and the blurring characteristics of the syste...
详细信息
ISBN:
(纸本)9781510673311;9781510673304
This work focuses on enhancing the quality of A- and B-scans of a novel linear optical coherence tomography system (LOCT), addressing the image degradation caused by noise and the blurring characteristics of the system's three-dimensional point spread function. The enhancement procedure includes an initial spatial and frequency-based pre-filtering that is applied to the measured interference pattern. Subsequently, a more robust envelope detection technique based on the Hilbert transform is employed. Lastly, image structures are reconstructed using a deconvolution algorithm based on maximum likelihood estimation, tailored to meet our unique requirements by adapting it to Rician distributed intensity values and employing a sparseness regularization term. For the deconvolution, both the lateral and axial blur of the system are considered. Emphasis is placed on the optimization of signal detection in high-noise regions, while simultaneously preventing image boundary artifacts. The efficacy of this approach is demonstrated across multiple types of measurement objects, including both artificial and biological samples. All results show a significant reduction in noise as well as enhanced resolution. Structure distinguishability is also increased, which plays a crucial role in tomography applications. In summary, the proposed enhancement method substantially improves image quality. This is achieved by still using the same initial measurement data, but incorporating prior knowledge and maximizing the amount of extracted information. Although initially designed for LOCT systems, the processing steps have potential for broader application in other types of optical coherence tomography and imaging systems.
Face morphing attacks have posed severe threats to Face Recognition systems (FRS), which are operated in border control and passport issuance use cases. Correspondingly, morphing attack detection algorithms (MAD) are ...
详细信息
ISBN:
(纸本)9798350365474
Face morphing attacks have posed severe threats to Face Recognition systems (FRS), which are operated in border control and passport issuance use cases. Correspondingly, morphing attack detection algorithms (MAD) are needed to defend against such attacks. MAD approaches must be robust enough to handle unknown attacks in an open-set scenario where attacks can originate from various morphing generation algorithms, post-processing and the diversity of printers/scanners. The problem of generalization is further pronounced when the detection has to be made on a single suspected image. In this paper, we propose a generalized single-image-based MAD (S-MAD) algorithm by learning the encoding from Vision Transformer (ViT) architecture. Compared to CNN-based architectures, ViT model has the advantage on integrating local and global information and hence can be suitable to detect the morphing traces widely distributed among the face region. Extensive experiments are carried out on face morphing datasets generated using publicly available FRGC face datasets. Several state-of-the-art (SOTA) MAD algorithms, including representative ones that have been publicly evaluated, have been selected and benchmarked with our ViT-based approach. Obtained results demonstrate the improved detection performance of the proposed S-MAD method on inter-dataset testing (when different data is used for training and testing) and comparable performance on intra-dataset testing (when the same data is used for training and testing) experimental protocol.
暂无评论