Re: Image sensor OV9655 and

[email protected] Wed, 26 Jan 2011 22:19:37 +0100
Newsgroups gmane.comp.video.video4linux
Message-ID <[email protected]>
Paolo,

In order to capture a portion of an image (ROI), both sensor and  =

PXA270 Quick Capture Interface must be correctly configured. I don't  =

know OV9655 sensor but you can find similar solution in the video  =

drivers for MBS270 V2 board:
http://www.mobisensesystems.com/fics/xscale_BSPs/2009_08/system/src/linux-2=
.6.26-mbs270.tar.bz2

Best regards,

Florian PANTALEAO


Paolo Santinelli <[email protected]> a =E9crit=A0:

> Dear Sirs,
> Hi all,
>
> I am currently trying to write C code for an embedded smart camera
> equipped with an XScal-PXA270 processor running Linux 2.6.9-dma270 and
> the OV9655 Image sensor.
>
> I am asking help, my goal is capture a subfield of the whole image.
>
> How to do it?
>
> Is this function (VIDIOCSCAPTURE or something else?) supported by the
> Video4Linux driver for the OV9655 Image sensor ?
>
> Currently I am able to reconfigure the OV9655 in order to change the
> whole image resolution, 640x480, 320x240 and even lower. The problem
> is I need do it dynamically, quickly, without close the camera and
> then reopen it every time I need sub-capture the image.
>
> I don't have any experience using device driver.
>
> I have tried to change the capture area using this call:
>
> vw.width  =3D new_image_width;
> vw.height =3D new_image_height;
>
> ioctl(fd_camera, VIDIOCSWIN, &vw)
>
> where vw is a struct video_window.
>
> I get the right image but it seems to be not vertically centred, it
> appears shifted and wrapped vertically. To get the right image I have
> to close and reopen the camera. The  closing and opening operations
> take too much time. I would like dynamically  change the  linux device
> driver behaviour without close and reopen the camera.
>
> BTW, the VIDIOCSCAPTURE doesn't work!
>
> Somebody could help me ?
>
> Thanks
>
> Paolo Santinelli
>
> --
> --------------------------------------------------
> PhD student Paolo Santinelli
> ImageLab Computer Vision and Pattern Recognition Lab
> Dipartimento di Ingegneria dell'Informazione
> Universita' di Modena e Reggio Emilia
> via Vignolese 905/B, 41125, Modena, Italy
>
> Cell. +39 3472953357,=A0 Office +39 059 2056270, Fax +39 059 2056129
> email:=A0 <mailto:[email protected]> [email protected]
> URL:=A0 <http://imagelab.ing.unimo.it/> http://imagelab.ing.unimo.it
> --------------------------------------------------
>
> --
> video4linux-list mailing list
> Unsubscribe mailto:[email protected]?subject=3Dunsubscr=
ibe
> https://www.redhat.com/mailman/listinfo/video4linux-list
>





--
video4linux-list mailing list
Unsubscribe mailto:[email protected]?subject=3Dunsubscribe
https://www.redhat.com/mailman/listinfo/video4linux-list