Posts

I really love the feeling about using photo to represent some themes. Recently, I tried lightroom, which is a post-process software for photos. I learned many color theorys and tools behind it. Let’s have a look. Color Theory. Additive Color: RGB for Red, Green and Blue. Red + Green + Blue = White Subtractive Color: These are created by mixing two additive colors. Red + Green = Yellow Green + Blue = Cyan Red + Blue = Magenta Yellow + Cyan + Magenta = Black Basics Histogram: a graph that displays the brightness and color distribution of an image From left to right, each column represents the number of the white-black pixels in different level It has five different region, from left to right, they are blacks, shadows, exposure, highlights and whites. There are two triangles at the top left and top right. the top left one is shadow clipping which will show you whether your image has region that was too dark the top right one is highlight clipping which show you whether your image has region that was too light the reason using clipping is because we can’t represent a value of a pixel by the number of bits. For example, the most common bit to represent image is 8 bits(0-255), such as (0，128， 254). When the value over 255 or less than 0, the pixel will be break, and the information will be clipped to 255 and 0. There are several colors in the triangle gray - all details are well preserved red - red channel overexposure/underexposure green - green channel overexposure/underexposure blue - blue channel overexposure/underexposure yellow - yellow channel overexposure/underexposure cyan - cyan channel overexposure/underexposure magenta - magenta channel overexposure/underexposure solid white - RGB channel channel overexposure/underexposure in the same time White balance Temperature Tint Tone Exposure: change the overall brightness of the photo Contrast: Increase or decrease the difference between highlights and shadows in a photo. It can make highlights brighter and shadows darker or vice versa. Highlights: Only the brighter areas in the photo are controlled, excluding the brightest pure white areas, such as cloud details in the sky, hightlights on the skins, etc. Shadows: Only the darker areas in the photo, but does not include the darkest pure black parts, such as the details of people in the shadows, the dark parts of buildings, etc. Whites: Define the brightest point in photo, it controls the rightmost end of the brightness range and determines which parts of your image become pure white. Blacks: Define the “darkest point” or “black point” of your photo. It controls the leftmost end of the brightness range and determines which parts of the image will become pure black. Presence Texture: Focus on the surface texture, such as skin pores, surface of rocks, etc. Clarity: 它不像“纹理”那么精细，而是让物体的轮廓和结构显得更“硬朗”或更“柔和”。It isn’t as fine-detailed as “Texture”. Instead, it makes the outlines and structures of objects appear “harder” or “softer”. Dehaze: 主要用于消除照片中的大气薄雾、雾霾或朦胧感，同时增加色彩的饱和度。It is primarily used to eliminate atmospheric haze, smog, or mist in a photo, while simultaneously increasing color saturation. Vibrance: 它会优先提升画面中本身不太饱和的颜色（比如天空的蓝色、植物的绿色），而对于已经很饱和的颜色则影响较小. It selectively boosts the less saturated colors in an image (like the blue in the sky or the green in plants), while having a smaller impact on colors that are already highly saturated. Saturation: 一个“简单粗暴的”色彩增强工具。它会对画面中的所有颜色进行无差别的、同等程度的提升。A “simple and heavy-handed” color enhancement tool. It boosts all colors in the image indiscriminately and to the same degree. Tone Curve The square graph looks similar with histogram, but contain a line segment from bottom left to top right. We can pull the curve to change tone in specific region, pull the curve left and up will make region lighter(add more RGB) and right down make region darker(add more CMYK) for point curve. we can also pull RGB channel separately. S curve to increase contrast Tips: e.g. to increase Yellow, we can pull the blue-yellow curve to yellow more, or we can pull red-cyan to red more and green-magenta to green more. This post will update frequently. ...

Recently, I tried Distrobox, a tool that enables us to run different operating systems within the host OS. I’ve noticed it’s different from a virtual machine, as it uses containerization instead. Over the past year, I’ve heard a lot about containers (like Docker), sandboxes, and virtual machines, and I’d like to understand the distinctions between them. Virtualization How virtualization work Virtualization is the technology that allows a single physical machine, known as the host, to run multiple virtual machines or guests. ...

The Global Interpreter Lock (GIL) At its core, the Global Interpreter Lock, or GIL, is a lock that only allows one thread to execute Python bytecode at a time within a single process. This means that even on a multi-core processor, a standard Python program with multiple threads will only utilize a single core for executing Python code. It is also for compatibility with large number extension modules written in C. These C extensions may not have built-in thread safety mechanisms, so GIL provides a safety net to ensure that they are executed in a single-threaded environment. The primary reason for the GIL’s existence lies in Python’s memory management. ...

Bitcoin The legend of Bitcoin has shown its magic for a long time. Recently, I have started to explore this field, and this is a record of my learning. The Blockchain⛓️ The blockchain is the foundational technology of Bitcoin. Think of it as a public, digital ledger or receipt book that is shared across thousands of computers worldwide. It’s a Chain of Blocks: Each “block” contains a list of recent transactions. When a new block is created, it is cryptographically linked to the previous one, forming an unbroken chain leading all the way back to the very first block. It’s Immutable: Because each block is linked to the one before it, changing a transaction in an old block would require re-doing all the work for every single block that came after it. This makes the ledger permanent and tamper-proof. This structure is what proves each coin’s history and prevents fraud like double-spending. The Genesis Block📜 The very first block, known as the Genesis Block, was mined on 2009.01.04, by Bitcoin’s mysterious creator, Satoshi Nakamoto. This single block was the start of the entire Bitcoin network. Once it was created, the race to mine the second block began, and the chain has been growing continuously ever since. ...

2024 Nobel Physics prize was earned by Professor John Hopfield and Professor Geoffrey Hinton, to thanks their distribution on machine learning. However, I felt very suprised that why it gives to machine learning? Anyway, I hadn’t deeply find the answer in that time. Recently, Professor Geoffrey Hinton gived us a short lecture about Boltzmann machine virtually. And before the lecture, I learned Hopfield Network(the predecessor of Boltzimann machine) at my accommodation. So today I will record this moment. ...

Book - Unix: A History and a Memoir Recently, I read this fantastic book. It bring me back to that 1960s - a period without modern computer and how the most clever minds in this world changed the world. During the reading, I found many answers to the “why” questions I had when I learning linux system. AT&T built Bell Labs and invited some of the most brilliant people in the world to do the most advanced scientific work. There was no limit on funding and no fixed goals for individuals. The system developed before Unix was called Multics. Since “Multics” already used “multi,” the early name of Unix was “Unics.” Unix was first written on the PDP-7. The next version, written in C, was developed on the PDP-11. Fortunately, it wasn’t written for the PDP-10. Tools like the shell, grep, regular expressions, the C language, the C compiler, yacc, lex, make, sed, awk, and troff were all invented at Bell Labs. Unix eventually declined due to copyright issues. AT&T sold it as a product and made it proprietary, which gave rise to open-source Unix-like systems. GNU is a Unix-like project that provides free and open-source alternatives. Under the GNU license, if you modify the source code of a project, the modified version must also remain open-source. MacOS is based on BSD, which is a Unix-like system. The Linux kernel combined with GNU forms GNU/Linux. They both follow POSIX. In the early days, operating systems were not portable. This changed with the invention of the C language and its compiler. MINIX was widely used because it was embedded in Intel chips. The working environment at Bell Labs in the 1970s are of hard problems, brilliant colleagues with shared dreams, and a unique management style that encouraged innovation. Microsoft once had its 3own Unix-like system. Another completely different path from Unix was MS-DOS, which eventually evolved into today’s Windows. You can also get to know the geniuses of that era, like Ken Thompson, Richard Stallman, and Brian Kernighan. “Everything is a file” is one of the core principles of Unix. The KISS principle (“Keep It Simple, Stupid”) is a fundamental part of Unix philosophy. The UNIX philosophy is very similar to some programming concepts I’ve recently learned at university. That’s why, Its impact not only on system desisgn but also software deveopemnt and beyond. ...

Model Context Protocol overview MCP (Model Context Protocol) can be understood as a “universal language” for communication between AI and external tools. It’s like a translator, allowing different AI applications (such as chatbots, code assistants) and different tools (like databases, GitHub, calendars) to easily communicate without needing to develop a new interface every time. Why is MCP needed? In the past, if you wanted an AI assistant to access different tools, like a calendar, email, or task manager, you would need to develop a separate interface for each tool (function calling), which resulted in a huge amount of work (N AI applications × M tools = N×M interfaces). MCP simplifies everything: all AI applications only need to support MCP, and all tools only need to support MCP. This way, they can communicate with each other, reducing development costs (N+M interfaces). ...

Email sending and receiving system The main system is build based on three protocols: SMTP, POP3 and IMAP. SMTP is used for sending emails to the recipient’s email server, but it does not handle receiving emails. User1 sends an email via an email client, and the email is first sent to User1’s email server using SMTP. Then, the email server forwards it to the recipient’s email server using SMTP as well. ...

In this semester, I listened the course mit 6.006 in youtube channel. Duirng the course, the professor used different notations to represents time complexity of an algorithm. I learned Big O O(n) notation before, but for Big theta θ(n) and reccurence relations T(n), I never heard them before. Today, I hope I can finally figure them out. What T(n) represents the actual running time of an algorithm O(n) represents the asymptotic upper bound of the running time of an algorithm θ(n) represents the running time when asymptotic upper bound and lower bound of an algorithm ares the same. ...

This week, I browsed many old machines at ebay to use for my first attempt at setting up a homelab. Initially, I planned to build a machine myself during the summer holiday, but in today I found a great performance and a high cost-effective machine that changed my mind. Overview The machine model is HP-Elitedesk-800-G4-SFF. Compared to its previous generation, the chipset supports 8th and 9th generation of intel core cpu, which offers a significant improvement(6c6t) compare with 6th/7th core cpu(4c4t). Furthermore it provides NVme express in this generation. With these components, I can build a highly cost-effective homelab. The total cost is £150. ...