Author name cluster

Haibo Zhou

Possible papers associated with this exact author name in Arrow. This page groups case-insensitive exact name matches and is not a full identity disambiguation profile.

3 papers

1 author row

EAAI Journal 2024 Journal Article

Asymmetric convolutional multi-level attention network for micro-lens segmentation

Shunshun Zhong
Haibo Zhou
YiXiong Yan
Fan Zhang
Ji'an Duan

Tiny target recognition in automation is currently a hot research task that usually suffers from typical issues such as complex background, dim target, and slow detection speed. In the current study, a data-driven method is proposed to realize the posture recognition of micro-lens during optical device coupling to achieve accurate clamping of the gripper. First, we establish a pixel-by-pixel labeled optical micro-lens dataset named single-frame micro-lens target (SFMT), which provides data support for the subsequently proposed convolutional neural network. Subsequently, an asymmetric convolutional multi-level attention network (ACMANet) is proposed to realize accurate segmentation detection of micro-lenses by employing an embedded multi-scale asymmetric convolutional module (MACM) and a multi-level interactive attention module (MIAM). MACM achieves not only a reduction in computational complexity but also enhanced robustness for rotated image recognition through multi-scale asymmetric convolutional kernels. Furthermore, MIAM improves the accuracy of image segmentation by connecting the down-sampling and up-sampling stages and realizing the fusion of pixel position details and key channel features. Extensive experimental results based on our self-constructed image acquisition system demonstrate that the values of normalized intersection over union and dice are successively 91. 41% and 95. 50%, and the processing speed is 3. 3 s/100 images, which shows the advance of ACMANet.

Details DOI

EAAI Journal 2024 Journal Article

Derivation and characteristic investigation of a two-input, two-output interval type-2 fuzzy controller using product and operations

Song Yin
Haibo Zhou
Ji-an Duan
Zhenli Huang

Interval type-2 (IT2) fuzzy logic, with the footprint of uncertainty (FOU) in membership functions, is very effective in dealing with disturbances and uncertainties of nonlinear systems. With this advantage, a two-input, two-output IT2 fuzzy controller with IT2 input and singleton type-1 fuzzy output sets is proposed for coupling systems. Firstly, the controller's input-output relationship is determined. Derivation results indicate that the controller is equivalent to the sum of a relay module and a local proportional-integral (PI) controller with variable gains. Next, analytical structure properties and gain variations are explored. It is demonstrated that the relay module is independent of the FOU, playing a dominant role in determining the control behavior of the IT2 controller. On the other hand, the control action of the PI controller diminishes when increasing the FOU and the number of fuzzy input sets. These properties contribute to the fast responsiveness of the controller and advantages in handling system uncertainty. Furthermore, the stability condition of this controller is established using the small gain theorem. Finally, both simulation and experiments are carried out to illustrate the applicability of the IT2 controller. The experimental results show that the controller's performance is embodied as the tracking errors of the two-linkage axes equal to ±4 μm, while the repeated positioning accuracy values are within ±3 μm.

Details DOI

NeurIPS Conference 2019 Conference Paper

Graph-Based Semi-Supervised Learning with Non-ignorable Non-response

Fan Zhou
Tengfei Li
Haibo Zhou
Hongtu Zhu
Ye Jieping

Graph-based semi-supervised learning is a very powerful tool in classification tasks, while in most existing literature the labelled nodes are assumed to be randomly sampled. When the labelling status depends on the unobserved node response, ignoring the missingness can lead to significant estimation bias and handicap the classifiers. This situation is called non-ignorable non-response. To solve the problem, we propose a Graph-based joint model with Non-ignorable Non-response (GNN), followed by a joint inverse weighting estimation procedure incorporated with sampling imputation approach. Our method is proved to outperform some state-of-art models in both regression and classification problems, by simulations and real analysis on the Cora dataset.

PDF Details