Zhejiang Institute of Metrology

The global textile industry, challenged by resource depletion and environmental pollution, urgently requires a shift toward a circular economy. However, recycling efforts remain limited, focusing mainly on conventional fibers and neglecting high-performance materials like aramid. This study addresses the recycling of used firefighting protective clothing-an aramid-rich, high-turnover waste stream typically landfilled or incinerated. Life cycle assessment reveals the significant carbon footprint of its production and disposal, underscoring the need for circular strategies. A systematic recycling framework is established, integrating collection logistics and redesign principles. A graded "three-tier" approach is proposed, enabling direct reuse, yarn regeneration, and non-woven production based on material conditions. High-value products were developed by incorporating firefighting heritage and intangible cultural crafts into the design, supported by digital product passports for traceability. These strategies enhanced market acceptance and emotional value. The work provides a scalable circular solution for high-performance textiles, aiming to extend material life, reduce carbon emissions, and advance sustainable textile management through a novel combination of technical recycling and cultural value addition.

In order to ensure the transmission of diagnostic - level dose measurement values and the development and testing of novel diagnostic X - ray measurement devices, we established a reference radiation field in strict accordance with the standards of IEC 61267 and YY/T 0481. Subsequently, we conducted measurements on the performance of this reference radiation field and defined the RQR, RQA, and RQT series of radiation qualities. The results demonstrate that the diameter of the uniform radiation field exceeds 100 mm. The X - ray half - value layer meets the standard requirements, with the measured half - value layer falling within the range of 0.485 - 0.515, and the stability being less than 1%. These results fully satisfy the criteria for a standard reference radiation field. The establishment of the reference radiation field has effectively resolved the issue of value traceability for diagnostic - level dosimeters in Zhejiang Province. Moreover, it has furnished a crucial testing platform for medical radiation diagnostic measurement equipment, thereby facilitating the advancement of relevant research and development work in this field.

Abstract The CMOS temperature sensor, with its advantages of low power consumption, high integration, and digital output, has been widely adopted in smart healthcare, IoT, and wearable devices. To address performance variations caused by process deviations during mass production, this study proposed an efficient multi-temperature-point calibration method. The calibration system employed standard platinum resistance thermometers, thermal chambers, and precision electrical measurement equipment. The stability was calculated using Bessel’s formula, confirming the repeatability of the calibration methodology. Experimental results demonstrated that calibrated sensors achieved significantly reduced measurement errors within the -55 °C to 215 °C range, with stability performance better than 0.41 °C. The research proved that both standard voltage and current inputs could enable rapid calibration, providing an effective solution for reducing production costs and enhancing product consistency.

Small object detection is a critical task in applications like autonomous driving and ship black smoke detection. While Deformable DETR has advanced small object detection, it faces limitations due to its reliance on CNNs for feature extraction, which restricts global context understanding and results in suboptimal feature representation. Additionally, it struggles with detecting small objects that occupy only a few pixels due to significant size disparities. To overcome these challenges, we propose the Context-Aware Enhanced Feature Refinement Deformable DETR, an improved Deformable DETR network. Our approach introduces Mask Attention in the backbone to improve feature extraction while effectively suppressing irrelevant background information. Furthermore, we propose a Context-Aware Enhanced Feature Refinement Encoder to address the issue of small objects with limited pixel representation. Experimental results demonstrate that our method outperforms the baseline, achieving a 2.1% improvement in mAP.

Reliable visual perception is essential for autonomous driving test scenario generation, yet adverse weather and lighting variations pose significant challenges to simulation robustness and generalization. Traditional unpaired image-to-image translation methods primarily rely on RGB-based transformations, often resulting in geometric distortions and loss of structural consistency, which can negatively impact the realism and accuracy of generated test scenarios. To address these limitations, we propose a Depth-Aware Dual-Branch Generative Adversarial Network (DAB-GAN) that explicitly incorporates depth information to preserve spatial structures during scenario generation. The dual-branch generator processes both RGB and depth inputs, ensuring geometric fidelity, while a self-attention mechanism enhances spatial dependencies and local detail refinement. This enables the creation of realistic and structure-preserving test environments that are crucial for evaluating autonomous driving perception systems, especially under adverse weather conditions. Experimental results demonstrate that DAB-GAN outperforms existing unpaired image-to-image translation methods, achieving superior visual fidelity and maintaining depth-aware structural integrity. This approach provides a robust framework for generating diverse and challenging test scenarios, enhancing the development and validation of autonomous driving systems under various real-world conditions.

Abstract In computer vision, Person Re-Identification (ReID) is essential for surveillance, matching images of individuals across different cameras and lighting conditions. Despite progress in visible light scenarios using CNNs, methods struggle in low-light, prompting the need for Visible-Infrared (VI) ReID. The challenge lies in bridging the semantic gap between visible and infrared modalities due to their distinct imaging properties. Existing methods are limited, focusing on complex feature extractors without fully utilizing intermediate representations to reduce cross-modality discrepancies. Infrared images lack color information, and differences in reflectivity between modalities hinder the extraction of robust features for identification. To alleviate these problems, We introduce ML-RID, leveraging Mutual Learning (ML) to enhance the performance of the VI-ReID task. Specifically , we propose an Adaptive Feature Fusion Module (AFFM) that dynamically fuses visible and infrared features, compensating for semantic deficiencies and enhancing feature representation. Then, a Feature Projection Module (FPM) aligns predictions across modalities, while the Jensen-Shannon Divergence is utilized to ensure the prediction consistency and reliability, encouraging learning from each modality’s strengths. A penal term is proposed to maintain auxiliary modality diversity, aiding in knowledge transfer as a “teacher”. Experiments on two datasets show ML-RID outperforms current models, with visualizations verifying its effectiveness, marking a step forward in cross-modality Re-ID.

Freeform surfaces that can offer high degrees of freedom for aberration correction have been widely used in various imaging applications. Initial design is particularly critical in the optical design of freeform imaging systems due to the significantly expanded solution space. Here, we present a method to find an initial point for the optical design of four-mirror freeform imaging systems. The method is computationally simple, easily accessible, and theoretically supported. The effectiveness of the method is demonstrated by designing a four-mirror confocal system without field-constant aberration and linear astigmatism. We also generalize the proposed method to four-mirror confocal systems with “double-pass surface.” The proposed method can yield a promising starting point for the design of freeform off-axis imaging systems.

Abstract Freeform liquid–crystal polarization imaging optics (LPIOs) have vast potential for application in next‐generation interactive displays like augmented reality and virtual reality. However, tailoring the phase profile of a freeform LPIO for flexible aberration correction and diffraction efficiency control is still a big challenge due to the difficulties arising from the non‐zero étendue of an imaging system and the finite bandwidth of LPIOs. Here, a practical approach to creating freeform LPIOs from design to fabrication is proposed. A freeform polarization exposure is developed to produce a flexible, precise, and ultra‐high‐resolution photoalignment pattern for a freeform LPIO. A cooperative optimization is performed between the freeform LPIO and the freeform polarization exposure system to find the optimal phase of the freeform LPIO and the optimal surface profiles of the freeform lenses employed in the exposure system to produce freeform wavefronts. The optimal phase of the freeform LPIO is then achieved by superposition interference of the engineered freeform wavefronts via freeform polarization exposure. The effectiveness of the approach and the advantages of freeform LPIOs is experimentally demonstrated. The proposed approach will significantly promote the widespread applications of freeform LPIOs in the field of optics and photonics.