# Calibrated images not well positioned in raycloud (step 1) **URL:** \ **Category:** PIX4Dmapper\ **Created:** [September 27, 2026, 6:41pm UTC](https://community.pix4d.com/t/calibrated-images-not-well-positioned-in-raycloud-step-1/34709 "2026-09-27T18:41:28Z")\ **Posts on this page:** 5\ **Page:** 1
**Author:** ![amdubois](https://avatars.discourse-cdn.com/v4/letter/a/a88e4f/32.png) [@amdubois](https://community.pix4d.com/u/amdubois)\ **Post date:** [September 27, 2026, 6:41pm UTC](https://community.pix4d.com/t/calibrated-images-not-well-positioned-in-raycloud-step-1/34709/1 "2026-09-27T18:41:28Z")
Hi, I’m running into a weird issue. I’ve processed this 2 flight dataset of multispectral images. My step 1 results are great over all but I’ve noticed 5 images that are completely offset, although calibrated. ![image](https://global.discourse-cdn.com/pix4d/original/2X/8/8b0b8b2fc5ac6b1f5de78e8746a6fd4f8b2ae711.jpeg) First, I’m wondering: why? Second, I’m wondering how come this does not seem to have an influenced on my results. (see report attached) Third, should I deactivate them and reoptimize (or rematch and optimize)? Kind regards for your support! a.marie [Johnville\_Nord\_ms\_report.pdf](https://community.pix4d.com/uploads/short-url/xnpIKyJftGxxR242cU5wquHc4hk.pdf) (2.0 MB) ---
**Author:** ![Kapil\_Khanal](https://sea2.discourse-cdn.com/pix4d/user_avatar/community.pix4d.com/kapil_khanal/32/304_2.png) [@Kapil\_Khanal](https://community.pix4d.com/u/Kapil_Khanal)\ **Post date:** [September 29, 2026, 7:10pm UTC](https://community.pix4d.com/t/calibrated-images-not-well-positioned-in-raycloud-step-1/34709/2 "2026-09-29T19:10:33Z")
Hello @amdubois, Those miss positioning of the images can be due to bad calibration. I can see few issue on the keypoint matches. ![image](https://global.discourse-cdn.com/pix4d/original/2X/9/96751da43469e0ba4396d96950a8076e109d7d6c.jpeg) I would recommend changing the calibration setting as below. 1. Calibration Method to “ **Alternative** ”. ---
**Author:** ![amdubois](https://avatars.discourse-cdn.com/v4/letter/a/a88e4f/32.png) [@amdubois](https://community.pix4d.com/u/amdubois)\ **Post date:** [October 1, 2026, 12:32am UTC](https://community.pix4d.com/t/calibrated-images-not-well-positioned-in-raycloud-step-1/34709/3 "2026-10-01T00:32:23Z")
Hello @Kapil_Khanal , Alternative calibration has already been tested by another colleague on this same dataset. Although he used All for Internal parameter optimization, it yielded even worst results. (See attached report). I can’t say however if he had misplaced calibrated cameras since it is not available anymore. [Johnville\_Nord\_report.pdf](https://community.pix4d.com/uploads/short-url/2Z9VYWFtXrXw34N2nZkhro8cUCu.pdf) (2.4 MB) ---
**Author:** ![Kapil\_Khanal](https://sea2.discourse-cdn.com/pix4d/user_avatar/community.pix4d.com/kapil_khanal/32/304_2.png) [@Kapil\_Khanal](https://community.pix4d.com/u/Kapil_Khanal)\ **Post date:** [October 1, 2026, 1:30pm UTC](https://community.pix4d.com/t/calibrated-images-not-well-positioned-in-raycloud-step-1/34709/4 "2026-10-01T13:30:15Z")
Hello @amdubois, In general, if you dataset is not working with standard template then that is the indication that your dataset it not good. If might be things like overlap, missing of the images, blurred images, wrong geotag and so on. So maybe it would be good to see what can be improved in your image acquisitions. I would say, remove those unwanted images as shown in the screenshot below because it is not adding any value but might cause issues. If your images orientation are not good, just remove it and process without it. Inspect the images in the area you are having issues. ![image](https://global.discourse-cdn.com/pix4d/original/2X/a/a598242f39e2f42043201cd3bb27add21d4dabeb.png) ---
**Author:** ![amdubois](https://avatars.discourse-cdn.com/v4/letter/a/a88e4f/32.png) [@amdubois](https://community.pix4d.com/u/amdubois)\ **Post date:** [October 1, 2026, 2:14pm UTC](https://community.pix4d.com/t/calibrated-images-not-well-positioned-in-raycloud-step-1/34709/5 "2026-10-01T14:14:21Z")
Unfortunately, acquisition parameters for this flight were not ideal I must say. It was the first flight with a new drone/new sensors (we got lidar and rgb on top of ms sensor). The person that designed the flight plan worked with shutter interval time instead of direct overlap values and since speed has a direct effect on the results, I calculate that we are roughly around 90% percent longitudinal but only 70% lateral. Lessons have been learned. That being said, the take off/return images were removed in the second process if you look at the first shared report (Johnville\_Nord\_ms\_report). But from what you said, I understand I shouldn’t keep those wrongly calibrated images. **So how come the software displays them as good though?** I would except some kind of warning or at least another colour in the raycloud to show me that, although those images were calibrated, they clearly are way off from the excepted values. Betterm the software should invite the user to choose whether or not she/he wants to deactivate them before reoptimizing. Maybe food for thoughts on futur improvements? Thanks again @Kapil_Khanal