Horizon detection algorithm [closed]

Closed. This question needs to be more focused. It is not currently accepting answers.

---

Want to improve this question? Update the question so it focuses on one problem only by editing this post.

Closed 7 years ago.

Improve this question 开发者_JS百科

Can you help me find any information on how to detect horizon on image?

It should not be based on genetic algorithm or neural network.

---

Just found this question interesting, so I searched the internet for you and came up with following papers/links, the first one perhaps being the most interesting, as it provides a concrete algorithm.

1. Towards Flight Autonomy: Vision-Based Horizon Detection for Micro Air Vehicles (PDF at citeseer)
2. Attitude Estimation for a Fixed-Wing Aircraft Using Horizon Detection and Optical Flow (PDF)

Following the citations in the papers you will get to more resources on research in this field.

---

I'm not sure does this works. But my first approach would be - to detect most frequent line by using Hough transform with such properties:

• line should extend up to image boundaries.
• line divides image into two regions such that in one of them standard deviation of color is small.

---

Following procedure will detect horizon:-

1. Change RGB image to grayscale.
2. Find the edges in the image using Canny edge detector. Adjust the sigma of gaussian filter.
3. Apply hough transform on it.
4. Select the line segment with highest value of J (Equation 5) in Towards Flight Autonomy: Vision-Based Horizon Detection for Micro Air Vehicles