Zobrazeno 1 - 10
of 533
pro vyhledávání: '"multi-scale feature extraction"'
Publikováno v:
Complex & Intelligent Systems, Vol 10, Iss 6, Pp 8095-8108 (2024)
Abstract Prohibited item detection is crucial for the safety of public places. Deep learning, one of the mainstream methods in prohibited item detection tasks, has shown superior performance far beyond traditional prohibited item detection methods. H
Externí odkaz:
https://doaj.org/article/d484803ea4e94f1398b877c6caf694a2
Publikováno v:
Scientific Reports, Vol 14, Iss 1, Pp 1-13 (2024)
Abstract UNet architecture has achieved great success in medical image segmentation applications. However, these models still encounter several challenges. One is the loss of pixel-level information caused by multiple down-sampling steps. Additionall
Externí odkaz:
https://doaj.org/article/ff06928ac7a043fea32ee08b15754a2c
Publikováno v:
Scientific Reports, Vol 14, Iss 1, Pp 1-14 (2024)
Abstract In low-light environments, the amount of light captured by the camera sensor is reduced, resulting in lower image brightness. This makes it difficult to recognize or completely lose details in the image, which affects subsequent processing o
Externí odkaz:
https://doaj.org/article/e0c20eca30a24106a8b3c0bfba15b566
Publikováno v:
Mathematical Biosciences and Engineering, Vol 21, Iss 4, Pp 5007-5031 (2024)
In demanding application scenarios such as clinical psychotherapy and criminal interrogation, the accurate recognition of micro-expressions is of utmost importance but poses significant challenges. One of the main difficulties lies in effectively cap
Externí odkaz:
https://doaj.org/article/bb37ac35d7de4c80af658865a4809bca
Publikováno v:
Electronic Research Archive, Vol 32, Iss 4, Pp 2267-2285 (2024)
Nowadays, advancements in facial recognition technology necessitate robust solutions to address challenges in real-world scenarios, including lighting variations and facial position discrepancies. We introduce a novel deep neural network framework th
Externí odkaz:
https://doaj.org/article/e6581bac6cda405190e71be1e5a97143
Publikováno v:
Ain Shams Engineering Journal, Vol 15, Iss 8, Pp 102861- (2024)
Single image dehazing is a fundamental but challenging task in image processing. Various deep learning-based methods have achieved great dehazing performance. However, there are still hazy residues, even color distortion and texture loss when removin
Externí odkaz:
https://doaj.org/article/21c8262f14b340269d82495f1ab11005
Autor:
Ning Zhang, Wenqing Zhu
Publikováno v:
Frontiers in Physics, Vol 12 (2024)
The disparity between human and machine perception of spatial information presents a challenge for machines to accurately sense their surroundings and improve target detection performance. Cross-modal data fusion emerges as a potential solution to en
Externí odkaz:
https://doaj.org/article/76e1430ffc3a4334a1000451da6b5717
Publikováno v:
IEEE Access, Vol 12, Pp 104300-104316 (2024)
Panoramic image object detection has significant applications in autonomous driving, robotic navigation, and security monitoring. However, most current object detection algorithms are trained on pinhole images and cannot be directly applied to panora
Externí odkaz:
https://doaj.org/article/5a6d6aee743a4f9abc22b58507a5a2e5
Autor:
Tianao Chen, Aotian Chen
Publikováno v:
IEEE Access, Vol 12, Pp 4544-4560 (2024)
Deep learning has shown superiority in change detection (CD) tasks, notably the Transformer architecture with its self-attention mechanism, capturing long-range dependencies and outperforming traditional models. This capability provides the Transform
Externí odkaz:
https://doaj.org/article/c62cf6a9bd434144bfa62af560bd21f2
Autor:
Pengtao Lv, Heliang Xu, Yana Zhang, Qinghui Zhang, Quan Pan, Yao Qin, Youyang Chen, Dengke Cao, Jingping Wang, Mengya Zhang, Cong Chen
Publikováno v:
Insects, Vol 15, Iss 11, p 827 (2024)
In the process of rice production, rice pests are one of the main factors that cause rice yield reduction. To implement prevention and control measures, it is necessary to accurately identify the types of rice pests and diseases. However, the applica
Externí odkaz:
https://doaj.org/article/6bbfd018b7044f918924b1571b4eb3a7