.png){kind=icon}
返回
Enhancing representational learning for cloud robotic vision through explainable fuzzy convolutional autoencoder framework
.png){kind=icon}
PRE
AI
DOI：10.1007/s00500-023-08570-6.png)
.png){kind=icon}
原文求助.png){kind=icon}
分享.png){kind=icon}
收藏
摘要
En 中文
Robot vision is one of the most recent developments in robotics and automation technology. Owing to machine vision systems that integrate image processing and deep learning, robots could now operate faster on the assembly line and in new contexts such as supermarkets, hospitals, and restaurants. The key driver for such systems is the advancement of machine vision systems. Robotic vision is mainly composed of programs, cameras, and any other technology that assists robots in developing visual insights. This enables robots to do sophisticated visual tasks, such as a robot arm trained to pick up an object. The conventional algorithms employ semantic segmentation from camera images. Reconstructing a real-time environment from two or more images entails creating a 3D model of the scene require expensive labeled data to train visual semantic segmentation. These 3D models can be anything from fragments of a 3D point cloud to 3D surface models produced using advanced techniques. In order to extract several images into 3D models or subsets of a raw point cloud, enhanced representation learning is necessary. Inspired by the recent achievements of machine vision and deep learning, this paper proposes a deep representation learning technique for enhanced feature recognition learning based on explainable convolutional autoencoder that can be employed for feature classifier level using fuzzy clustering and further facilitated the human decision making of data, which requires for the representation of robot data that allow for swift and precise observations.
Keyword:
Deep learning
Convolutional autoencoder
Robot vision
Feature recognition
3D modeling
Robotics
Computer vision
期刊
IF:
2.5
论文数:
1.0W
被引数:
2.1W
机构
引用论文
YOLO-Tomato: A Robust Algorithm for Tomato Detection Based on YOLOv3Yolo-tomato: 一种基于YOLOv3的鲁棒番茄检测算法.png)
SENSORS
OAAI
SENSORSIF3.5
OAAI分享收藏Study of MEH–PPV/PCBM active layer morphology and its application for hybrid solar cell performanceMEH–PPV/PCBM活性层形貌及其在混合太阳能电池性能中的应用研究OAAI
OAAI分享收藏Development of Chlorhexidine Loaded Halloysite Nanotube Based Experimental Resin Composite with Enhanced Physico-Mechanical and Biological Properties for Dental Applications开发用于牙科应用的具有增强的物理机械和生物学性能的氯己定负载埃洛石纳米管基实验树脂复合材料OAAI
OAAI分享收藏分享收藏Artificial Intelligence-Enabled Sensing Technologies in the 5G/Internet of Things Era: From Virtual Reality/Augmented Reality to the Digital Twin5g/物联网时代的人工智能传感技术: 从虚拟现实/增强现实到数字孪生OAAI
OAAI分享收藏分享收藏
查看更多内容