Astrophotography

月食时月面边缘有一条青绿色的窄带，科普说那是臭氧吸收。但为什么是窄带不是整圈？为什么全食最深时反而看不见它？我们从最土的白圆盘开始，一层层加物理，硬算出这条带，在一路翻车里发现了一个更大的问题：AI 太懂物理，反而会把你带进前人留下的近似里。

During a lunar eclipse, a narrow green-blue band appears at the moon's edge; popular science says it is ozone absorption. But why a narrow band instead of a full ring? Why does it vanish at deepest totality? We start from the simplest white disk, layer in physics, and compute the band — and through repeated failures discover a bigger problem: AI knows physics too well, and will happily lead you into approximations left behind by earlier researchers.

从最直白的"一星一像素"出发，八次翻车、六亿颗星，一步一步把银河从真实星表里逼出来。在这个过程中才发现，以前从来没认真想过头顶的星空为什么长这个样子。

Starting from the simplest "one star, one pixel" rendering, eight failures and 600 million stars later, the Milky Way slowly emerged from a real star catalog.

探索用生成式AI解决天文图像中CMOS灰尘去除难题。通过让AI生成灰尘遮罩而非直接修复图像，结合确定性数学平场校准，实现既保留科学严谨性又获得AI强大效果的工作流。

Explores using generative AI to remove CMOS dust from astronomical images. By having AI generate dust masks rather than direct corrections, then applying deterministic flat-field calibration, achieves both scientific integrity and AI's powerful recognition capabilities.

用Skyfield库模拟行星位置，通过可见性、直线排列度和黄经集中度三项指标量化评估行星连珠现象，实现天体排列的定量分析与预测。

A quantitative approach to identifying planetary alignments using orbital mechanics and Skyfield library, combining visibility, linear alignment, and longitude clustering metrics to search decades of astronomical data.

5分钟用AI给全天相机加上云量识别功能：利用GPT-4V理解云量，让AI自动编写调用API的代码，展示AI时代开发效率的革命性提升。

天文摄影自动对焦算法思考：用信息熵代替星点拟合描述对焦质量，加权曲线拟合优先考虑合焦点附近数据，实现稳定精确的自动对焦。