Currently I’m testing Dahua IPC-HFW7842H-Z. Hopefully the review will be ready soon, but there is still plenty to do, and I have to find time to put all material together. Anyway, while testing the camera, I’ve found something I was thinking about since my first Dahua I bought. That is, how to get the best quality image from the camera, omitting all video compression. I was wondering if this is possible at all.
It turns out, it is. Unfortunately it only works with snapshots, not video files. Normally, when you use motion detection, the camera records a video, compressed with h264/h265 codec. If you turn on snapshot recording, the still images will be also compressed with this compression and then saved as JPG files. In other words, it will be only a frame grabbed from already compressed video file. Same goes for snapshot function while watching live footage through Web Interface. If you click snapshot button, you’ll get a large – apparently uncompressed - bitmap file, but all the compression will be already there.
What I’ve noticed though, is that when using IVS smart functions, the camera is pushing data straight from the sensor, omitting all video compression. The only compression applied is JPG and it is controlled by snapshot quality setting in VIDEO settings. Snapshots taken by IVS functions are completely independent of video compression you set. You can have the bitrate set to 1000kbps and while the video will look awfully, the snapshot will be good. Look at the example below:
On the left side is a frame grabbed from video file. The bitrate was set to 7000kbps. For a 4K camera it is not much. On the right side you see a snapshot taken with a use of Video Structuralization function. The difference is obvious. I’ve checked this on my other cameras, and it worked everywhere. The Tripwire function will do the trick.
It turns out, it is. Unfortunately it only works with snapshots, not video files. Normally, when you use motion detection, the camera records a video, compressed with h264/h265 codec. If you turn on snapshot recording, the still images will be also compressed with this compression and then saved as JPG files. In other words, it will be only a frame grabbed from already compressed video file. Same goes for snapshot function while watching live footage through Web Interface. If you click snapshot button, you’ll get a large – apparently uncompressed - bitmap file, but all the compression will be already there.
What I’ve noticed though, is that when using IVS smart functions, the camera is pushing data straight from the sensor, omitting all video compression. The only compression applied is JPG and it is controlled by snapshot quality setting in VIDEO settings. Snapshots taken by IVS functions are completely independent of video compression you set. You can have the bitrate set to 1000kbps and while the video will look awfully, the snapshot will be good. Look at the example below:
On the left side is a frame grabbed from video file. The bitrate was set to 7000kbps. For a 4K camera it is not much. On the right side you see a snapshot taken with a use of Video Structuralization function. The difference is obvious. I’ve checked this on my other cameras, and it worked everywhere. The Tripwire function will do the trick.
