I’m wondering if anyone else is experiencing this. Every so often, I get a photo with an orange tint (35 out of 80 on my last mission although sometimes it is none or only one or two). If there are only a couple, I disable them from the project prior to processing, but sometimes there are too many to do this. I’m flying with the P2V+ and using the Android App. I’ve never experienced this when taking photos manually with the DJI app.
Due to the IOS issues, I just purchased an android tablet, I managed to get one good test flight, but when, I switched to camera, I also noticed that the onscreen image looked yellowish. Actual images appeared to be fine though, unlike yours being captured that way. The second flight with android version was total disaster. Mission start, switched to camera, app crashed, restarted app, phantom still flying mission, but not taking images. Finished mission, went into hover at last point on flight line, had to use atti, to get a hold of phantom and land it. It did not return and land. No images.
Sorry to hear that your second Android app was a bust. I would try again. I’ve flown over 150 missions with the Android app and have maybe had it crash a handful of times. I’m using a Samsung Galaxy Tab 4 if that makes any difference. Also, I never switch to the camera view, so that might be the issue?
As for the orange glow on the photos, it is actually on the captured images. Even if I take the SD card out of the Phantom the photos are orange. And it is happening on more than one Phantom. Seems strange if it is only me.
Weather has been a little moist here for a few days, tablet is a branded AT&T with lollipop. Specs seemed pretty good, not the best but good. I wiped it and reinstalled everything, reinstalled capture, hoping that will fix. I will be sure to post back , if I run into captured images showing up yellowish or orange. Forecast for next few days is bleak. May be a bit, before I can fly.
I have experienced the same issue after running the android app for the first time yesterday. Unlike your experience I have gotten those mis colored images directly from just the DJI Vision App. I was curious to see if I would get the same result from Mapper, and I did. I thought I had an issue with my camera.
We have also noticed the same issue with some test missions we performed, some images appear more yellowish or orange. It seems to be a white balance issue.
In the application, we set the camera white balance to auto but for some reason, it seems that the camera is not always handling correctly the auto setting.
We have noticed that this problem mostly arises when it is very sunny. If you have this problem you can try to fly when the sky is homogeneously covered by cloud but not too dark or too cloudy.
However, this should not affect the reconstruction in Pix4Dmapper but it can affect the visual aspect of the orthomosaic.
We are working together with DJI in order to investigate this issue.
We are also considering giving the user the option to select between different lightening conditions such as cloudy, sunny, etc. in a future version of the application.
This would be excellent if we could select the weather type before flying. It is not always possible to wait for perfect conditions before flying.
And no, the orange tint does not affect the stitching at all, just makes weird orange blotches on the orthomosaic.
Glad to hear that it is not just me anyways. Thanks.
I have the same problem where I get 5 or 6 images that have a very warm cast and the next 5 or 6 images turn blue, I experienced the problem first time when I was using the dji app, now it happened with the Pix4D beta app. Interesting to note in the forum that this happens mostly in bright sunlight, I will do a test using an ND filter and see if I get the same problem.
Would be nice to have manual control of the white balance in the Pix4D app. Auto tends to have issues in full sun when there’s lots of dry grass in the photo.
I am also experiencing this problem. Indeed, it mainly occurs when it is very sunny both with the Pix4D capture app and with the DJI go app. I’ve tried to use white balance options Auto, Sunny and Cloudy but the difference of white balance between pictures remains.
Are there already solutions regarding this problem?
A small update on this matter.
We are aware of white balance (WB) issues with DJI cameras reported by some users. In the same dataset, images can appear hotter or cooler with different shades. This depends on the temperature of the scene that is captured, meaning its content (different local sources of light) and the outdoor or indoor light conditions.
The WB parameter allows to compensate temperatures so that the color of the image is as close as possible to neutral which corresponds to sunlight at noon outdoor. At the moment, the white balance is set to automatic when flying an autopilot mission with our app. For each image the temperature is adjusted by an algorithm that optimizes the colors. We are considering to make available template lighting conditions that the user could select. A WB setting was already implemented on iOS (Auto, Sunny, Cloud) but not on Android.
The problem is that in very rare cases, the DJI camera does not handle correctly the Auto WB and even the Sunny or Cloud iOS options might not be sufficient to ensure homogeneous temperature among the image set. However, it does not affect the reconstruction in Pix4Dmapper but it can affect the visual aspect of the orthomosaic and the colors of the points in the rayCloud.
The issue usually arises when the mapped area is particularly hot like earth or desert, meaning the colors are warmer than usually and with little contrast (few cool shades). Also changing light conditions because of clouds for instance, is difficult to handle by the camera. The best is to fly at noon with a sunny weather to ensure stable and constant lightning.
Ideally we would like DJI to add a tool to their SDKs that allows to set the correct color temperature to all images from the same flight. They said they might consider it in the future.
Hope this will help to better understand the situation.