arXiv subjects

Qing Chang

Publications and source records attributed to Qing Chang.

28 records · Page 2Linked to original sources

arXiv2024

UltraLight VM-UNet: Parallel Vision Mamba Significantly Reduces Parameters for Skin Lesion Segmentation

Traditionally for improving the segmentation performance of models, most approaches prefer to use adding more complex modules. And this is not suitable for the medical field, especially for mobile medical devices, where computationally loaded models are not suitable for real clinical environments due to computational resource constraints. Recently, state-space models (SSMs), represented by Mamba, have become a strong competitor to traditional CNNs and Transformers. In this paper, we deeply explore the key elements of parameter influence in Mamba and propose an UltraLight Vision Mamba UNet (UltraLight VM-UNet) based on this. Specifically, we propose a method for processing features in parallel Vision Mamba, named PVM Layer, which achieves excellent performance with the lowest computational load while keeping the overall number of processing channels constant. We conducted comparisons and ablation experiments with several state-of-the-art lightweight models on three skin lesion public datasets and demonstrated that the UltraLight VM-UNet exhibits the same strong performance competitiveness with parameters of only 0.049M and GFLOPs of 0.060. In addition, this study deeply explores the key elements of parameter influence in Mamba, which will lay a theoretical foundation for Mamba to possibly become a new mainstream module for lightweighting in the future. The code is available from https://github.com/wurenkai/UltraLight-VM-UNet .

eess.IV↗

arXiv2024

H-vmunet: High-order Vision Mamba UNet for Medical Image Segmentation

In the field of medical image segmentation, variant models based on Convolutional Neural Networks (CNNs) and Visual Transformers (ViTs) as the base modules have been very widely developed and applied. However, CNNs are often limited in their ability to deal with long sequences of information, while the low sensitivity of ViTs to local feature information and the problem of secondary computational complexity limit their development. Recently, the emergence of state-space models (SSMs), especially 2D-selective-scan (SS2D), has had an impact on the longtime dominance of traditional CNNs and ViTs as the foundational modules of visual neural networks. In this paper, we extend the adaptability of SS2D by proposing a High-order Vision Mamba UNet (H-vmunet) for medical image segmentation. Among them, the proposed High-order 2D-selective-scan (H-SS2D) progressively reduces the introduction of redundant information during SS2D operations through higher-order interactions. In addition, the proposed Local-SS2D module improves the learning ability of local features of SS2D at each order of interaction. We conducted comparison and ablation experiments on three publicly available medical image datasets (ISIC2017, Spleen, and CVC-ClinicDB), and the results all demonstrate the strong competitiveness of H-vmunet in medical image segmentation tasks. The code is available from https://github.com/wurenkai/H-vmunet .

cs.CV↗

arXiv2023

Only Positive Cases: 5-fold High-order Attention Interaction Model for Skin Segmentation Derived Classification

Computer-aided diagnosis of skin diseases is an important tool. However, the interpretability of computer-aided diagnosis is currently poor. Dermatologists and patients cannot intuitively understand the learning and prediction process of neural networks, which will lead to a decrease in the credibility of computer-aided diagnosis. In addition, traditional methods need to be trained using negative samples in order to predict the presence or absence of a lesion, but medical data is often in short supply. In this paper, we propose a multiple high-order attention interaction model (MHA-UNet) for use in a highly explainable skin lesion segmentation task. MHA-UNet is able to obtain the presence or absence of a lesion by explainable reasoning without the need for training on negative samples. Specifically, we propose a high-order attention interaction mechanism that introduces squeeze attention to a higher level for feature attention. In addition, a multiple high-order attention interaction (MHAblock) module is proposed by combining the different features of different orders. For classifying the presence or absence of lesions, we conducted classification experiments on several publicly available datasets in the absence of negative samples, based on explainable reasoning about the interaction of 5 attention orders of MHAblock. The highest positive detection rate obtained from the experiments was 81.0% and the highest negative detection rate was 83.5%. For segmentation experiments, comparison experiments of the proposed method with 13 medical segmentation models and external validation experiments with 8 state-of-the-art models in three public datasets and our clinical dataset demonstrate the state-of-the-art performance of our model. The code is available from https://github.com/wurenkai/MHA-UNet.

cs.CV↗

arXiv2023

Generalized Kernel Regularized Least Squares

Kernel Regularized Least Squares (KRLS) is a popular method for flexibly estimating models that may have complex relationships between variables. However, its usefulness to many researchers is limited for two reasons. First, existing approaches are inflexible and do not allow KRLS to be combined with theoretically-motivated extensions such as random effects, unregularized fixed effects, or non-Gaussian outcomes. Second, estimation is extremely computationally intensive for even modestly sized datasets. Our paper addresses both concerns by introducing generalized KRLS (gKRLS). We note that KRLS can be re-formulated as a hierarchical model thereby allowing easy inference and modular model construction where KRLS can be used alongside random effects, splines, and unregularized fixed effects. Computationally, we also implement random sketching to dramatically accelerate estimation while incurring a limited penalty in estimation quality. We demonstrate that gKRLS can be fit on datasets with tens of thousands of observations in under one minute. Further, state-of-the-art techniques that require fitting the model over a dozen times (e.g. meta-learners) can be estimated quickly.

stat.ML↗

arXiv2022

Reinforcement Learning Based User-Guided Motion Planning for Human-Robot Collaboration

Robots are good at performing repetitive tasks in modern manufacturing industries. However, robot motions are mostly planned and preprogrammed with a notable lack of adaptivity to task changes. Even for slightly changed tasks, the whole system must be reprogrammed by robotics experts. Therefore, it is highly desirable to have a flexible motion planning method, with which robots can adapt to specific task changes in unstructured environments, such as production systems or warehouses, with little or no intervention from non-expert personnel. In this paper, we propose a user-guided motion planning algorithm in combination with the reinforcement learning (RL) method to enable robots automatically generate their motion plans for new tasks by learning from a few kinesthetic human demonstrations. To achieve adaptive motion plans for a specific application environment, e.g., desk assembly or warehouse loading/unloading, a library is built by abstracting features of common human demonstrated tasks. The definition of semantical similarity between features in the library and features of a new task is proposed and further used to construct the reward function in RL. The RL policy can automatically generate motion plans for a new task if it determines that new task constraints can be satisfied with the current library and request additional human demonstrations. Multiple experiments conducted on common tasks and scenarios demonstrate that the proposed user-guided RL-assisted motion planning method is effective.

eess.SY↗

arXiv2022

Paint shop vehicle sequencing based on quantum computing considering color changeover and painting quality

As customer demands become increasingly diverse, the colors and styles of vehicles offered by automotive companies have also grown substantially. It poses great challenges to design and management of automotive manufacturing system, among which is the proper sequencing of vehicles in everyday operation of the paint shop. With typically hundreds of vehicles in one shift, the paint shop sequencing problem is intractable in classical computing. In this paper, we propose to solve a general paint shop sequencing problem using state-of-the-art quantum computing algorithms. Most existing works are solely focused on reducing color changeover costs, i.e., costs incurred by different colors between consecutive vehicles. This work reveals that different sequencing of vehicles also significantly affects the quality performance of the painting process. We use a machine learning model pretrained on historical data to predict the probability of painting defect. The problem is formulated as a combinational optimization problem with two cost components, i.e., color changeover cost and repair cost. The problem is further converted to a quantum optimization problem and solved with Quantum Approximation Optimization Algorithm (QAOA). As a matter of fact, current quantum computers are still limited in accuracy and scalability. However, with a simplified case study, we demonstrate how the classic sequencing problem in paint shop can be formulated and solved using quantum computing and demonstrate the potential of quantum computing in solving real problems in manufacturing systems.

quant-ph↗

arXiv2022

Fiber spectrum analyzer based on planar waveguide array aligned to a camera without lens

We propose and experimentally demonstrate a fiber spectrum analyzer based on a planar waveguide chip butt-coupled with an input fiber and aligned to a standard camera without any free-space optical elements. The chip consists of a single-mode waveguide to connect with the fiber, a beam broadening area, and a waveguide array in which the lengths of the waveguides are designed for both wavelength separation and beam focusing. The facet of the chip is diced open so that the outputs of the array form a near-field emitter. The far field are calculated by the Rayleigh-Sommerfeld diffraction integral. We show that the chip can provide a focal depth on the millimeter scale, allowing relaxed alignment to the camera without any fine-positioning stage. Two devices with 120 and 220 waveguides are fabricated on the polymer waveguide platform. The measured spectral width are 0.63 nm and 0.42 nm, respectively. This simple and practical approach may lead to the development of a spectrum analyzer for fiber that is easily mountable to any commercial camera, thereby avoiding the complication for customized detectors as well as electronic circuits afterwards.

physics.optics↗

arXiv2022

DATA: Domain-Aware and Task-Aware Self-supervised Learning

The paradigm of training models on massive data without label through self-supervised learning (SSL) and finetuning on many downstream tasks has become a trend recently. However, due to the high training costs and the unconsciousness of downstream usages, most self-supervised learning methods lack the capability to correspond to the diversities of downstream scenarios, as there are various data domains, different vision tasks and latency constraints on models. Neural architecture search (NAS) is one universally acknowledged fashion to conquer the issues above, but applying NAS on SSL seems impossible as there is no label or metric provided for judging model selection. In this paper, we present DATA, a simple yet effective NAS approach specialized for SSL that provides Domain-Aware and Task-Aware pre-training. Specifically, we (i) train a supernet which could be deemed as a set of millions of networks covering a wide range of model scales without any label, (ii) propose a flexible searching mechanism compatible with SSL that enables finding networks of different computation costs, for various downstream vision tasks and data domains without explicit metric provided. Instantiated With MoCo v2, our method achieves promising results across a wide range of computation costs on downstream tasks, including image classification, object detection and semantic segmentation. DATA is orthogonal to most existing SSL methods and endows them the ability of customization on downstream needs. Extensive experiments on other SSL methods demonstrate the generalizability of the proposed method. Code is released at https://github.com/GAIA-vision/GAIA-ssl

cs.CV↗

arXiv2020

Mastering the working sequence in human-robot collaborative assembly based on reinforcement learning

A long-standing goal of the Human-Robot Collaboration (HRC) in manufacturing systems is to increase the collaborative working efficiency. In line with the trend of Industry 4.0 to build up the smart manufacturing system, the Co-robot in the HRC system deserves better designing to be more self-organized and to find the superhuman proficiency by self-learning. Inspired by the impressive machine learning algorithms developed by Google Deep Mind like Alphago Zero, in this paper, the human-robot collaborative assembly working process is formatted into a chessboard and the selection of moves in the chessboard is used to analogize the decision making by both human and robot in the HRC assembly working process. To obtain the optimal policy of working sequence to maximize the working efficiency, the robot is trained with a self-play algorithm based on reinforcement learning, without guidance or domain knowledge beyond game rules. A neural network is also trained to predict the distribution of the priority of move selections and whether a working sequence is the one resulting in the maximum of the HRC efficiency. An adjustable desk assembly is used to demonstrate the proposed HRC assembly algorithm and its efficiency.

cs.RO↗

arXiv2018

Switchable plasmonic scattering of nanogaps for linewidth-tunable random lasing

Linewidth-tunable lasers have great application requirements in the fields of high-resolution spectroscopy, optical communications and other industry and scientific research. Here, the switchable plasmonic scattering of the metal particles with plenty of nanogaps is proposed as an effective method to achieve linewidth-tunable random lasers. By using the nonlinear optical effect of the environment medium, the metal particles demonstrate the transition from local scattering of nanogaps with high spatial frequency to traditional Mie scattering free from detail information with increasing the pump power density. Based on these two scattering processes, random lasers can be continuously driven from a narrow-linewidth configuration exhibiting nanogap effect dominated resonances to a broad-linewidth regime of collectively coupling oscillating among nanowires (or nanoflowers), demonstrating the dynamic range of linewidth exceeds two orders of magnitude. This phenomenon may provide a platform for further studying of the conclusive mechanism of random lasing and supply a new approach to tune the linewidth of random lasers for further applications in high-illumination imaging and biology detection.

physics.optics↗