Closed. This question does not meet Stack Overflow guidelines. It is not currently accepting answers.
We don’t allow questions seeking recommendations for books, tools, software libraries, and more. You can edit the question so it can be answered with facts and citations.
Closed last month.
Improve this question
I know nothing on the subject of deep learning.
I am looking for references to build a deep learning algorithm to detect ROI in given images. My goal is to compare deep learning algorithms with usual image processing algorithms I have already made.
The input images look like this :
The output of the algorithm should look like this :
Q1: Do you have any references that if I read them would let me build such a deep learning algorithm from start to finish ?
Q2: Otherwise, do such algorithms already exist and are freely available ? (Note: Such algorithms should produce precise ROI detection not broad rectangles encircling the bright regions).
You can try using Mask R-CNN. Refer to these links for your understanding:
https://github.com/matterport/Mask_RCNN
https://arxiv.org/abs/1703.06870
Basically, you need to make an annotation (polygonal) for your dataset with tools like VIA image annotation tool (https://www.robots.ox.ac.uk/~vgg/software/via/) or MakeSense (https://www.makesense.ai/). These are the open source tools that I can recommend. After training, the network can predict the bounding box as well as the boundary of the detected objects.
Your task is easy, don't worry about knowing anything about the topic, as your image shows what you are trying to achieve, I would suggest you try using semantic segmentation, you can search on youtube or read about Faster R-CNNthey are kinda related to what you want to do. Then you can compare the output results with the regular image processing.
Related
Closed. This question does not meet Stack Overflow guidelines. It is not currently accepting answers.
We don’t allow questions seeking recommendations for books, tools, software libraries, and more. You can edit the question so it can be answered with facts and citations.
Closed 5 years ago.
Improve this question
I am hoping that TensorFlow can turn this input, to this output.
Input: A floorplan PNG, and 1 - 5 images of a symbol
Output: The same floorplan, but with all matching symbols highlighted
I can do the hard work of figuring out HOW to do it, but I don't want to waste 2 weeks just to figure out it wouldn't be possible. I know I'd need to train it with multiple images, but I won't have more than 5 examples of a given symbol.
Does TensorFlow have these capabilities?
Thanks!
Yes, it is possible to use tensorflow to create a machine learning algorithm to do that for you, but I would bet that is not how you want to do this. First off, in order to do this in tensorflow, you would need to manually create a large number of training samples and spend a significant amount of time figuring out how to define the network and train it. Sure, you could do it, but I definitely wouldn't advise it.
If you have a specific set of symbols that you want to highlight, it would probably be better to use opencv to find and highlight the symbols. For example, in opencv, you could use Template Matching to find a specific symbol in the floor plan and then highlight them by modifying pixel color.
Closed. This question does not meet Stack Overflow guidelines. It is not currently accepting answers.
We don’t allow questions seeking recommendations for books, tools, software libraries, and more. You can edit the question so it can be answered with facts and citations.
Closed 6 years ago.
Improve this question
Any tools could help recognize the data distribution pattern, and then make the decision to choose ML algorithms?
Firstly, you have to understand Machine Learning as a field, and have some understanding of its sub fields. If you don't intuitively understand your tools, you won't be able to identify when to use them.
The idea you're talking about is called exploratory data analysis, and it can be very approachable if you think about it the right way. Think about it in terms of the scientific method:
First, look over the data, and any documentation about it.
Then, come to some hypotheses about the patterns that might exist.
Based on your understanding of ML, brainstorm some approaches that might give some insight into your hypotheses. For example, if you see that your proposed dependent value can have several distinct values, you have a classification problem, and based on your input data, you should choose an appropriate approach.
The tools that you might find useful are plentiful, but a good start could be the programming language R, or Python. Both are very strong data science tools. R has a greater learning curve, but is built with data science in mind. Python, on the other hand, is very easy to pick up, but you have more choices to make with regards to ML and data science libraries. With Python, look into Pandas for CSV and data manipulation, and Tensorflow, Theano or Scikit-Learn for data analysis and ML.
Hope this helps!
Closed. This question does not meet Stack Overflow guidelines. It is not currently accepting answers.
We don’t allow questions seeking recommendations for books, tools, software libraries, and more. You can edit the question so it can be answered with facts and citations.
Closed 2 years ago.
Improve this question
I have a problem related to computer vision and machine learning. Basically we are working on video surveillance system which will be trained to detect any suspicious activity like theft or shop lifting in stores.We are confused that is that will be able to solve this problem or not. We don't know that is it feasible or not. So kindly just suggest us something. Any help will be appreciated.
While I understand that Open CV is great for face-detection and usable for face-recognition, can it be used for analyzing "actions", s.a. the act of sitting, the act of lifting an object off the shelf ? If so, what are some of these algorithms I should dig deeper into ?
Are there other libraries (other than OpenCV) which need to be used for such tasks? Are there open-source libraries for the same?
What you are trying to achieve is currently a very active area in computer vision and machine learning research called Behaviour Analysis or Activity Detection. State of the art approaches can be found in journals like PAMI or conferences like CVPR or NIPS. As of today, it is nowhere near the performance you would require to build an automatic theft-detection system in the general case (i.e., any surveillance camera looking into any scene in any orientation). Behaviour Analysis is based on many underlying techniques, such as identifying the pose of people in images. Current research is still trying to figure out if there's a person in the picture and the position of its limbs in the general case.
Here's what might be feasible with the current research state: A system that help an operator focus on potential threats when cameras have a clear unobstructed view to a clear and mostly static environment (e.g., glass displays). An operator could therefore monitor many more cameras than before, because the system will automatically hide the cameras that clearly does not contain suspicious activity or movement.
To know more about current possibilities, I recommend you to check the literature (like this example), decompose the problem into subparts and leverage your priors (your a priori knowledge of the scene and people you're looking at) as much as possible.
By using object recognition (by helping deep learning) we can detect object and by using the data set of recorded object in the shop we can assess to the detailed (price) of that object. based on the number of objects and information about the object we can recognize the issue such as thrift in the counter.
Closed. This question does not meet Stack Overflow guidelines. It is not currently accepting answers.
We don’t allow questions seeking recommendations for books, tools, software libraries, and more. You can edit the question so it can be answered with facts and citations.
Closed 7 years ago.
Improve this question
use for drone quadcopter to track human body
This problem depend on many factors :
Computational resources.
Quality of images.
How much accuracy do you expect from the algorithm
By the way, the easiest way for implementing such algorithm is Cascade Classifier which is implemented in OpenCV. You can train your own model or you can use the trained model which exists in openCV files. This method support three feature types: HOG,LBP and HAAR. The base of this method is paper Viola and Jones published on 2001. The test time is near to online in an ordinary computer.
If you need more accurate method you can try DPM (deformable part models) based method. There are many released version of this method on the internet. The speed of detection is almost 2 HZ.
If you need more accuracy I suggest you to go forward with CNN (Convolutional Neural Networks). Of Course you need more computational resources (GPU or high spec CPUs)
Closed. This question does not meet Stack Overflow guidelines. It is not currently accepting answers.
We don’t allow questions seeking recommendations for books, tools, software libraries, and more. You can edit the question so it can be answered with facts and citations.
Closed 7 years ago.
Improve this question
I am very beginner in robotics. I want to make a robotics project based on slam algorithms. I know many algorithm and i have the confidence to implement it in any language but i dont have any idea based on image processing and hardware. So, can anyone give a tuotorial based on slam based robotics projects[including how hardware organized and how image processing is done for that project], after seeing that i can make a slam based robotics project from my own.
In addition, If anyone give me a video lecture series for that then it would be very helpful.
Thanks in advance.
I have tried to do something similar last year. I created two systems. The first system made use of a camera and laser to detect objects and determine their location relative to the system itself. The second system was a little robot with tracks (wheels would be better), that used dead reckoning to keep track of its own location relative to its starting location. The techniques worked really well, but unfortunately I did not have the time to combine the two systems. I can however provide you with some documentation that was incredibly useful for me at that time.
These tutorials provide information on both the hardware and the software.
Optical Triangulation (detection of objects with a camera and laser) :
http://www.seattlerobotics.org/encoder/200110/vision.htm
Dead Reckoning (a technique to keep track of one's own location) :
http://www.seattlerobotics.org/encoder/200010/dead_reckoning_article.html