The Palace Museum (Chinese: 故宫博物院; pinyin: Gùgōng Bówùyùan) is a national museum housed in the Forbidden City at the core of Beijing. It was established in 1925 after the last Emperor of China was evicted from his palace, and opened its doors to the public.Constructed from 1406 to 1420, the museum consists of 980 buildings and covers 72 hectares (over 180 acres). It is home to over 1.8 million pieces of art, mostly from the imperial collection of the Ming and Qing dynasties. The 20th century saw its expansion through new acquisitions, transfers from other museums, and new archaeological discoveries.According to the Beijing Evening Post, the museum has seen more than 17 million visitors in 2018, which would make it the world's most visited museum. It has an average of 15 million visitors annually since 2012. Due to this increased pressure, the management has set a daily limit for visitors of 80,000 since 2015 to protect the structure and the experience.
Vision Large Language Models (VLLMs) have achieved remarkable success in modern text-rich visual understanding. However, their perceptual robustness in the face of the continuous morphological evolution of historical writing systems remains largely unexplored. Existing ancient text datasets typically focus on isolated historical periods, failing to capture the systematic visual distribution shifts spanning thousands of years. To bridge this gap and empower Digital Humanities, we introduce Chronicles-OCR, the first comprehensive benchmark specifically designed to evaluate the cross-temporal visual perception capabilities of VLLMs across the complete evolutionary trajectory of Chinese characters, known as the Seven Chinese Scripts. Curated in collaboration with top-tier institutional domain experts, the dataset comprises 2,800 strictly balanced images encompassing highly diverse physical media, ranging from tortoise shells to paper-based calligraphy. To accommodate the drastic morphological and topological variations across different historical stages, we propose a novel Stage-Adaptive Annotation Paradigm. Based on this, Chronicles-OCR formulates four rigorous quantitative tasks: cross-period character spotting, fine-grained archaic character recognition via visual referring, ancient text parsing, and script classification. By isolating visual perception from semantic reasoning, Chronicles-OCR provides an authoritative platform to expose the limitations of current VLLMs, paving the way for robust, evolution-aware historical text perception. Chronicles-OCR is publicly available at https://github.com/VirtualLUOUCAS/Chronicles-OCR.
For conservation and restoration purposes, the multi-analytical methods including optical microscopy (OM), scanning electron microscopy (SEM), Fourier transform infrared spectroscopy (FTIR) and thermally assisted hydrolysis-methylation pyrolysis-gas chromatography/mass spectrometry (THM-Py-GC/MS) were applied to characterize and identify the lacquered gauze fragment excavated from Han Tomb M59 in Lianying village, Yangzhou city, Jiangsu Province, China, dated back to 127-154 BC. The results showed that the lacquered gauze was made from tough and breathable ramie fabric as the base material and then the fabrics surface was treated with raw lacquer. When lacquering the fabrics, the linseed oil was added to modify the raw lacquer, improving the flexibility and gloss of the lacquer film. This study provides a scientific basis for the reconstruction of the lacquered gauze unearthed from the Han Dynasty, and helps to further understand the production materials and techniques of ancient lacquered gauze.
Eco-friendly substrates combining passive humidity regulation with visual indication remain limited for enclosed museum display cases. Poplar wood was modified by low-temperature ultrasonic co-impregnation with glycerol and anthocyanin. Substrate screening, single-factor tests, and orthogonal experiments quantified the effects of additive dosage and impregnation time on weight percent gain, dimensional stability, and humidity-responsive color change. SEM and FTIR indicated that modification was governed mainly by pore filling and intermolecular hydrogen bonding. The optimal process was 22 wt% glycerol, 0.6 wt% anthocyanin, and 9 h impregnation, yielding a 51% higher 24 h moisture uptake than untreated poplar, with effective pore filling and uniform pigment dispersion. The color response remained detectable after 30 wet-dry cycles. This work provides an environmentally compatible dual-functional wood substrate and a technical reference for passive humidity-buffering and visual monitoring in enclosed exhibition spaces, providing an effective tool for the preventive conservation of wooden heritage artifacts.
Virtual reconstruction of color-faded painting on bronze mirrors relies on limited information. Existing techniques face significant difficulties in the recognition and reconstruction of missing patterns at both structural and iconographic levels. Painting on bronze in burial environments is prone to decay, fading, contamination, and corrosion, leading to defective areas. Reconstruction of patterns is crucial for archeology and art history. Moreover, the complex and irregular patterns of painting pose a further challenge. A method is proposed to resolve these issues. Concretely, pigment types, particle sizes, painting stratigraphy and technique were analyzed by scanning electron microscopy with energy dispersive spectroscopy and Raman spectroscopy; these data were used to re-create the original color palette of the painting. The exact distribution of each pigment and the color patterns were ascertained by macroscopic X-ray fluorescence spectroscopy and hyperspectral imaging to re-create the line drawing. Digital coloring of the line drawing with palette values virtually reconstructs the original colors and appearance of the painting on the bronze mirrors under ideal conditions. The study demonstrates improvements in the faithful reproduction of the original colors on painted bronze and provides new perspectives for the study of painted bronzes. (c) 2025 Elsevier Masson SAS. All rights are reserved, including those for text and data mining, AI training, and similar technologies.
In this study, a picosecond laser-driven interfacial coalescence method is developed for realizing highly sensitive and uniform Au-based surface-enhanced Raman scattering (SERS) substrates via enhancement of aromatic ring vibrational modes, achieving an ultralow detection limit of 10(-18) M toward aromatic-ring-containing dyes and an exceptional enhancement factor of 1.83 & times;10(12). The confined environment restricts liquid Au flow, generating interfacial vortices within similar to 100 nm region and leading to the formation of high-density nanoparticles with excellent SERS performance. The enhancement mechanism is found to strongly correlate to the concentration, particularly at low concentration (<10(-8) M), the selective single-molecule-level enhancement of aromatic ring vibrational modes under sub-monolayer adsorption allows highly sensitive detection of five specific dyes and pigments with aromatic-ring (Rhodamine 6 G, Malachite Green, Alizarin Red, Carmine Red, and Indigo), confirming its broad applicability for trace molecular detection. The Au-based SERS substrate exhibits remarkable uniformity with a relative standard deviation of 5.74% and retains stable SERS activity for over 1 year under ambient conditions. This work establishes a scalable, cost-effective laser-based manufacturing paradigm for high-performance SERS sensors, offering exceptional sensitivity, reproducibility, and stability. The method enables ultralow-concentration detection of aromatic molecules via enhancement of aromatic ring vibrational modes, demonstrating versatility across cultural heritage preservation, environmental monitoring, and biomedical diagnostics.
