2014-09-29 11:57:52 -0600 | commented answer | Improve Brand Detection Quality I'm doing exactly in this way, and the results are these in question. |
2014-09-27 07:59:57 -0600 | commented answer | Improve Brand Detection Quality I followed your advice and I edited my question. My results is getting better with SURF than SIFT. It didn't solve the problem yet. |
2014-09-27 07:55:27 -0600 | received badge | ● Editor (source) |
2014-09-25 16:35:16 -0600 | asked a question | Improve Brand Detection Quality I am trying to detect brands in pictures but I'm having some problems. I did the cascade training with 500 positives samples and 1440 negatives samples, but I'm getting a lot of bad results. The cascade classifier detects the brand and further a lot of things that isn't the brand. EDIT: I tried to use SURF features to eliminate these bad results and look that it isn't work: The same result occured with a image that is not my brand: Look the detections: My question is: How do I eliminate the bad results and just stay with the brand that I am looking for ? What I'm doing wrong ? |
2014-09-08 09:14:12 -0600 | asked a question | Improve Cascade Detection I trained a cascade classifier with de following arguments: Positive Images: 500 Negative Images: 1440 Samples Creation: Train Cascade: I generated de cascade.xml, and I'm detecting in a group of images using OpenCV for Java with: My main goal is detect if there is or not the cascade classifier in searched images. So, finally my questions: 1 - What am I doing wrong ? 2 - Is there a good way to get the best detection results ? |
2014-08-05 21:39:34 -0600 | received badge | ● Supporter (source) |
2014-08-05 21:39:28 -0600 | commented answer | About Cascade Classifier Training It was very helpful for me, thanks! |
2014-08-05 20:42:51 -0600 | asked a question | Understanding OpenCV internally Hi, I want to know where I must to start studying OpenCV Algorithms internally. I have a good experience using OpenCV Library, but I want more. I know about the docs, but I find it a little confuse when explaining about math inside the Algorithms. I would like to know if you know books or some good content explaining OpenCV internally. Thanks! |
2013-10-31 22:02:38 -0600 | asked a question | About Cascade Classifier Training I'm collecting a lot of positive images to train the cascade classifier and so I have some doubts: 1 - I'm trying to recognize object that has 2 different formats, so it's better create 2 cascade's xml to each different format or 1 cascade's xml with a lot of 2 images formats ? 2 - The color of positive samples can influence in the object detection ? Is it better that the positive samples are in greyscale ? Thanks for your help ! |