define a scale on an image
显示 更早的评论
Hello, I'm using matlab gui to count objects in an image and plot an histogram which shows the size distribution of these latest. My problem is the X axis of the histogram corresponds to the number of pixels forming the edges of the objects. In the image i want to analyse, I have a bar which corresponds to the scale in meter. Does anyone know how to measure this scale to convert the number of pixels in the X axis to meter ?
Thank you in advance :) Hadi
回答（1 个）
Image Analyst
2017-2-9
0 个投票
You need to calculate/determine a spatial calibration factor. Are you looking through a microscope or telescope? The number of meters per pixel will be vastly different for those two cases. What is your field of view? Basically to get linear distances you need to multiply pixels by meters per pixel, and to get squared things, like area, you need to multiply by (meters per pixel)^2.
See my attached spatial calibration demo.
5 个评论
Hadi Hijazi
2017-2-9
Hello, thank you for your answer. This is the kind of images i have, and the scale is not the same for all my images, so i have to measure the scale in each one to know what is the scalar i should multiply by the number of pixels to obtain the real lengths of objects edges.
Thank you :)
Image Analyst
2017-2-9
You can easily find the length of the bar just by examining the line that contains the bar. But you'd have to know the real world distances. Hopefully it's in the header of the image and you don't have to do OCR on the numbers in the image.
Hadi Hijazi
2017-2-10
Thank you for your help :)
Nagendra babu Gurram
2018-11-8
can we use vanishing points for this problem
Image Analyst
2018-11-8
Yes but you'd have to measure at two locations and develop a spatial calibration "image" that gives you the spatial calibration at each location. I think that's probably what the camera calibration function in the Computer Vision System Toolbox does, though I haven't used it. See https://www.mathworks.com/products/computer-vision/features.html#camera-calibration since that's exactly what you need.
类别
在 帮助中心 和 File Exchange 中查找有关 Histograms 的更多信息
2017-2-8
2018-11-8
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!
Translated by 
