Zobrazeno 1 - 10
of 551
pro vyhledávání: '"residual connection"'
Publikováno v:
Scientific Reports, Vol 14, Iss 1, Pp 1-20 (2024)
Abstract Most existing scene classification methods based on remote sensing images tend to ignore important interactive information at different levels in the image. We propose an effective remote sensing scene classification method named multi-scale
Externí odkaz:
https://doaj.org/article/e1bd8ba7eff540a7b8bc8f609ff214e1
Publikováno v:
IET Image Processing, Vol 18, Iss 11, Pp 3028-3041 (2024)
Abstract Low‐light images are captured in environments with minimal lighting, such as nighttime or underwater conditions. These images often suffer from issues like low brightness, poor contrast, lack of detail, and overall darkness, significantly
Externí odkaz:
https://doaj.org/article/a61d2fe688554a4b9494bf0cefe8788d
Autor:
Xiaojun Bi, Yanlong Luo
Publikováno v:
Heritage Science, Vol 12, Iss 1, Pp 1-19 (2024)
Abstract Incomplete handwritten Dongba character often appears in heritage documents and its recognition is significant for heritage and philology. However, all previous methods always suppose that a complete Dongba character is used as input, and th
Externí odkaz:
https://doaj.org/article/e26858e28d5f45878f0d65c6b6bf0ff3
Publikováno v:
Frontiers in Bioengineering and Biotechnology, Vol 12 (2024)
Jaw cyst is a fluid-containing cystic lesion that can occur in any part of the jaw and cause facial swelling, dental lesions, jaw fractures, and other associated issues. Due to the diversity and complexity of jaw images, existing deep-learning method
Externí odkaz:
https://doaj.org/article/596305fc6b914137b19b9a0a1a756b20
Autor:
Chenlin Zhou, Han Zhang, Liutao Yu, Yumin Ye, Zhaokun Zhou, Liwei Huang, Zhengyu Ma, Xiaopeng Fan, Huihui Zhou, Yonghong Tian
Publikováno v:
Frontiers in Neuroscience, Vol 18 (2024)
Spiking neural networks (SNNs) offer a promising energy-efficient alternative to artificial neural networks (ANNs), in virtue of their high biological plausibility, rich spatial-temporal dynamics, and event-driven computation. The direct training alg
Externí odkaz:
https://doaj.org/article/175ed70e024c46d086d97f0923d8e960
Publikováno v:
Heritage Science, Vol 12, Iss 1, Pp 1-15 (2024)
Abstract To address the fuzzy segmentation boundaries, missing details, small target losses and low efficiency of traditional segmentation methods in ancient mural image segmentation scenarios, this paper proposes a mural segmentation model based on
Externí odkaz:
https://doaj.org/article/bd67cb6cdf0947409c1deae1d14a1398
Publikováno v:
IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, Vol 17, Pp 6003-6013 (2024)
High-resolution remote sensing imagery finds applications in diverse fields, such as land-use mapping, crop planning, and disaster surveillance. To offer detailed and precise insights, reconstructing edges, textures, and other features is crucial. De
Externí odkaz:
https://doaj.org/article/b5f28d97d3434a36a00fc17f7f5f7d31
Publikováno v:
IEEE Access, Vol 12, Pp 34553-34569 (2024)
Sign Language Recognition (SLR) represents a revolutionary technology aiming to establish communication between hearing impaired and non-hearing impaired communities, surpassing traditional interpreter-based approaches. Existing efforts in automatic
Externí odkaz:
https://doaj.org/article/66e878a09d2e46d49e588f4a699aefd1
Publikováno v:
IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, Vol 17, Pp 219-243 (2024)
Hyperspectral imagery is widely used for analyzing substances and objects, specifically focusing on their classification. The advancement of processing capabilities and the emergence of cloud computing platforms have made deep learning (DL) models in
Externí odkaz:
https://doaj.org/article/8c97ad2a060241b792acd3e7e43feafe
Publikováno v:
工程科学学报, Vol 45, Iss 11, Pp 1962-1969 (2023)
Aiming at the problems of insufficient feature expression ability and weak generalization ability of the traditional Res2Net model in the field of voice print recognition, this paper proposes a feature extraction module known as the SE-DR-Res2Block,
Externí odkaz:
https://doaj.org/article/62a63d64cd934bc39a2940f2ab491ed1