Invented by Wu; Zhe, Xu; Chuanxi, Younessian; Ehsan, DeLuccia; Luke, Braskich; Anthony

Welcome! Today, we’ll dive deep into an exciting patent application for smart camera systems that detect the state of openable things like doors and windows. We’ll break down the context, the science, and the inventive steps in a way that’s easy to understand—even if you aren’t a tech expert. Our goal is to help you see how this invention could change how we keep our homes and workplaces safe.

Background and Market Context

Security is a big concern for everyone. For years, people have used cameras and sensors to keep their homes and businesses safe. But traditional security systems often need a lot of hardware. You might have to place special sensors on every door or window. These systems can be expensive and hard to set up. Most of the time, they only tell you after someone has broken in or when something bad has already happened. They rarely warn you that, for example, a door was left open or a window was slightly ajar.

More and more, people want smarter solutions. They want their cameras not just to show what’s happening, but to understand it and send alerts when something is wrong—or even before something goes wrong. Maybe you want to know if your garage door is wide open while you’re away, or if your kitchen window was left open at night. Or maybe you want to ignore the alert if it’s just a tree blowing in the wind, not a real threat.

Smart cameras are now everywhere, from doorbells to baby monitors. But most of them still just show video or detect simple motion. They struggle to tell the difference between important events and false alarms, like pets moving around or curtains fluttering. And they rarely understand the difference between a door being open, closed, or just half open. This is where the new patent steps in.

This invention brings a fresh idea: using smart cameras to actually figure out the exact state of any openable structure—like a door, window, or gate—by looking at the picture itself. It promises to send you helpful alerts, based on what matters most to you, and not flood you with useless notifications. This is a big leap from old systems that need lots of sensors and wires. Now, with just a camera and smart software, you can know if your home is truly secure.

The market for these smart solutions is growing fast. People want easy-to-use technology that keeps them safe, without high costs or complicated installation. Businesses want smarter ways to keep track of their buildings. Homeowners want peace of mind, especially when they’re away. This invention, with its ability to use just a camera and smart processing, fits perfectly into this need. It offers a way to monitor and understand what’s happening in a space, in real time, with very little fuss.

Scientific Rationale and Prior Art

To really appreciate this patent, let’s look at how things work today and why this new approach is different.

In older security systems, cameras and special sensors work together. For example, a magnetic sensor on a door can tell if it’s open or closed. But each sensor has to be installed at the right spot, and you need one for every entry point. Over time, these sensors can fail or lose their calibration. And if something blocks the sensor, you might get a false reading. Plus, these systems often alert you only after a break-in, not before.

Motion detection in cameras is common. It works by comparing pictures taken every second. If something in the picture changes—say, a person walks by or a door moves—the camera sends an alert. The problem is, these alerts don’t tell you what actually changed. Was it the door opening, a person passing by, or just a shadow? You have to look at the video yourself to figure it out.

Some smart cameras try to detect objects, using what’s called “object recognition.” They might spot a person or a car. But recognizing whether a door is open or closed is much harder. The camera needs to learn what a door looks like in different positions, under different lighting, and with things in the way—like cars or pets. No two doors look exactly the same, and sometimes part of the door is hidden from view. This makes the problem challenging for computers.

More recent solutions use artificial intelligence (AI) and machine learning. These systems can learn from many images to spot patterns. But most of them still focus on detecting moving objects, not on understanding the state of a specific thing like a door or window. And if something blocks the camera’s view, the system might get confused.

Some companies have tried to use deep learning to analyze video feeds and recognize specific states. These methods often require huge amounts of data and lots of computer power. They may work well for some cases, but they don’t adapt easily when the scene changes—like when you move your camera or rearrange your furniture.

This patent builds on these ideas but goes further. It uses AI to not just detect motion or objects, but to find “regions of interest”—the important spots in the camera’s view, like where a door is. The system learns which parts of the picture matter most for telling if something is open or closed. It can ignore less important areas, like backgrounds or places that get blocked often.

The invention can also learn from history. It remembers what’s normal in your home—like the garage door is usually closed—and only alerts you when something unusual happens. It can even ask you for feedback, so it gets better over time. If you don’t want to get alerts about the front gate every time the wind blows, you can tell the system, and it will adjust.

By combining smart object detection, motion analysis, and machine learning, this patent claims a new way to understand what’s happening in your home or business. It can work with any camera—no need for extra wires or sensors. And it can adapt to your unique space, your doors and windows, and your needs.

Invention Description and Key Innovations

Let’s get to the heart of the invention. What does it actually do, and how does it do it?

At its core, the system uses one or more cameras to watch over a space. The camera sends pictures, or image data, to a computer—this can be a device in your home, or a remote server in the cloud. The computer looks at these pictures and tries to find “regions of interest.” These are the important parts of the scene, like where your door, window, or garage door is.

The system doesn’t just look for motion. It tries to understand what is in each part of the scene. For example, it can learn that there’s a door on the left side of the picture, and a window near the center. It can break up each region into smaller parts, or “weighted sections.” Some parts of the picture are more important than others for telling if the door is open or closed. For example, the bottom of a garage door might show the first sign of movement when the door opens. The system pays extra attention to these key areas.

Here’s how the process works, step by step:

First, the camera captures images of the space. The computer receives these images and looks for changes over time—like if pixels in the image become lighter or darker. If it sees a change, it marks that area as possibly important.

Next, the system decides which regions to focus on. It groups together pixels that move or change together—like the outline of a door. It can also use AI to recognize common things, like doors, windows, or gates. Sometimes, you can help the system by drawing a box around the door in the camera’s view, or by confirming what the system found.

Once the region of interest is set, the system breaks it into sections. Each section gets a “weight”—a number that tells how important that part is for deciding if the door is open or closed. The system learns these weights over time, by watching what happens and even by asking for your feedback. For example, if the lower right corner of the door always changes when the door moves, that section gets a higher weight.

The system keeps track of “values” for each section—like color, brightness, or other measurements. It compares these values to what it has seen before. If the values cross certain “thresholds,” the system decides the state has changed. For example, if the bottom of the garage door goes from dark to light, the system knows the door is open.

It can also look at history. If your front door is usually closed, but now it seems open, the system can alert you. It can even send messages to your phone, an app, or other devices. You can customize when you get alerts. For example, maybe you want to know if the door is open at night, but not during the day.

If you don’t want to be bothered by alerts every time your dog walks by the door, you can tell the system to ignore certain regions or types of motion. The system gets smarter as you use it. It can ask for your input—thumbs up or down on alerts—so it learns what matters to you.

Technically, the system uses advanced methods like neural networks and clustering to learn about your space. It can be trained on pictures of your home, learning what’s normal and what’s not. It can handle scenes where part of the door is blocked, like when your car is parked in front of the garage. It focuses on the visible sections that matter most.

The system is flexible. It can run some steps in your home, on a local device, and others in the cloud. It can work with many types of cameras. It can even adapt if you move the camera or change what’s in the room.

What makes this invention special is its ability to do all this without special hardware. Just a regular camera and smart software. It can tell not only if there’s motion, but what exactly changed and what it means. It can tell if a door is fully open, just cracked open, or closed tight. It can learn from you and get better over time, sending only the alerts you care about.

With this invention, you get safer spaces, fewer false alarms, and more control over your security. Businesses can monitor many doors and windows with fewer devices. Homeowners can relax, knowing they’ll be alerted if something is wrong, but not bothered by every small movement.

Conclusion

This patent application brings a big leap in home and business security. By using smart cameras and powerful software, it can truly understand what’s happening in your space. It helps you know if your doors or windows are open, closed, or somewhere in between—without the need for extra sensors or complicated setups. It adapts to your needs, learns from your feedback, and only alerts you when something really matters. As more people look for smart, easy-to-use ways to stay safe, inventions like this are leading the way. The future of security is smarter, simpler, and more personal—and this patent is a big step in that direction.

Click here https://ppubs.uspto.gov/pubwebapp/ and search 20250218261.