Extend the Vector class with compound assignment operators that match
the binary arithmetic operators.
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Milan Zamazal <mzamazal@redhat.com>
The Vector class defines multiple element-wise arithmetic operators
between vectors or between a vector and a scalar. A few variants are
missing. Add them.
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Milan Zamazal <mzamazal@redhat.com>
Instead of hand-coding all arithmetic operators, implement them based on
a generic apply() function that takes an operator-specific binary
operators. This will simplify adding missing arithmetic operators.
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Milan Zamazal <mzamazal@redhat.com>
The Vector class defines a set of arithmetic operators between two
vectors or a vector and a scalar. All the operators perform element-wise
operations, except for the operator*() that computes the dot product.
This is inconsistent and confusing. Replace the operator with a dot()
function.
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Milan Zamazal <mzamazal@redhat.com>
The default constructor leaves the vector data uninitialized. Add a
constructor to fill the vector with copies of a scalar value, and fix
the documentation of the default constructor.
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Milan Zamazal <mzamazal@redhat.com>
The Vector class can be useful to represent RGB pixel values. Add r(),
g() and b() accessors, similar to x(), y() and z(), along with an RGB
type that aliases Vector<T, 3>.
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Milan Zamazal <mzamazal@redhat.com>
The x(), y() and z() functions of the Vector class are convenience
accessors for the first, second and third element of the vector
respectively, meant to improve readability of class users when a vector
represents coordinates in 1D, 2D or 3D space. Those accessors are
limited to immutable access to the vector elements, as they return a
copy. Extend the API with mutable accessors.
The immutable accessors are modified to return a reference to the vector
elements instead of a copy for consistency. As they are inline
functions, this should make no difference in terms of performance as the
compiler can perform the same optimizations in their case.
While at it, reorder functions to declare operators before other member
functions, to be consistent with the usual coding style.
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Milan Zamazal <mzamazal@redhat.com>
The terms "shutter" and "shutter speed" are used through libcamera to
mean "exposure time". This is confusing, both due to "speed" being used
as "time" while it should be the inverse (i.e. a maximum speed should
correspond to the minimum time), and due to "shutter speed" and
"exposure time" being used in different places with the same meaning.
To improve clarity of the code base and the documentation, use "exposure
time" consistently to replace "shutter speed".
This rename highlighted another vocabulary issue in libcamera. The
ExposureModeHelper::splitExposure() function used to document that it
splits "exposure time into shutter time and gain". It has been reworded
to "split exposure into exposure time and gain". That is not entirely
satisfactory, as "exposure" has a defined meaning in photography (see
https://en.wikipedia.org/wiki/Exposure_(photography)) that is not
expressed as a duration. This issue if left to be addressed separately.
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Naushir Patuck <naush@raspberrypi.com>
Acked-by: Kieran Bingham <kieran.bingham@ideasonboard.com>
Report the default sensor black level reported by the datasheet.
Note that IMX327 and IMX462 depend on the IMX290 CameraSensorHelper.
That's fine since those sensors report the same defaults for the
black level as the Sony IMX290.
Signed-off-by: Geoffrey Van Landeghem <geoffrey.vl@gmail.com>
Reviewed-by: Jacopo Mondi <jacopo.mondi@ideasonboard.com>
Signed-off-by: Jacopo Mondi <jacopo.mondi@ideasonboard.com>
The sensor is largely compatible with the already supported
Sony IMX290 so we can reuse the same helpers for the analogue
gain conversion functions.
Signed-off-by: Geoffrey Van Landeghem <geoffrey.vl@gmail.com>
Reviewed-by: Jacopo Mondi <jacopo.mondi@ideasonboard.com>
Signed-off-by: Jacopo Mondi <jacopo.mondi@ideasonboard.com>
Some IPA modules, like the RkISP1 one, call FCQueue::get(0) at
IPA::start() time, before any frame context has been allocated with
FCQueue::alloc() called at queueRequest() time.
The FCQueue implementation aims to detect when a FrameContext is get()
before it is alloc()-ated, Warns about it, and initializes the
FrameContext before returning it.
In case of frame#0, a get() preceding an alloc() call is not detected
as the "frame == frameContext.frame" test returns success, as
FrameContexts are zeroed by default.
As a result, the first returned FrameContext is not initialized.
Explicitly test for frame#0 to make sure the FrameContext is initialized
if get(0) is called before alloc(0). To avoid re-initializing a frame
context, in case alloc() has been called correctly before get(),
introduce an "initialised" state variable that tracks the FrameContext
initialisation state.
Signed-off-by: Jacopo Mondi <jacopo.mondi@ideasonboard.com>
Reviewed-by: Daniel Scally <dan.scally@ideasonboard.com>
Reviewed-by: Stefan Klug <stefan.klug@ideasonboard.com>
We start to have some functions relating to colour that are
effectively identical crop up across the IPA modules. Add a file
allowing those to be centralised within libipa so that a single
implementation can be used in all of the IPAs.
Signed-off-by: Daniel Scally <dan.scally@ideasonboard.com>
Reviewed-by: Kieran Bingham <kieran.bingham@ideasonboard.com>
Reviewed-by: Milan Zamazal <mzamazal@redhat.com>
Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Previous iterations of the ExposureModeHelper class had a
setShutterGainLimits() function, which got renamed to setLimits(). The
documentation still uses the old name. Fix it.
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Jacopo Mondi <jacopo.mondi@ideasonboard.com>
Reviewed-by: Daniel Scally <dan.scally@ideasonboard.com>
The MatrixInterpolator is great for interpolation of matrices for
different color temperatures. It has however one limitation - it can
only handle matrices. For LSC it would be great to interpolate the LSC
tables (or even polynomials) using the same approach. Add a generic
Interpolator class based on the existing MatrixInterpolator. This class
can be adapted to any other type using partial template specialization.
Signed-off-by: Stefan Klug <stefan.klug@ideasonboard.com>
Reviewed-by: Kieran Bingham <kieran.bingham@ideasonboard.com>
Reviewed-by: Paul Elder <paul.elder@ideasonboard.com>
The libcamera coding style groups the C and C++ standard library headers
in a single group. Fix the few offenders in the source tree.
While at it, add a missing blank line between header groups in a
separate location.
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Milan Zamazal <mzamazal@redhat.com>
Reviewed-by: Daniel Scally <dan.scally@ideasonboard.com>
The documentation for the blackLevel_ member is very terse. Reference
the more complete documentation of the sibling blackLevel() member
function to provide more information.
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Milan Zamazal <mzamazal@redhat.com>
Reviewed-by: Kieran Bingham <kieran.bingham@ideasonboard.com>
Add black level value for OV5675 camera sensor.
According to datasheet, default value is 0x10, 10 bits width.
However, Linux kernel driver initializes black level target value
to 0x40. Set the value to the same as in kernel driver, but scaled
to 16 bits.
Signed-off-by: Daniel Semkowicz <dse@thaumatec.com>
Reviewed-by: Quentin Schulz <quentin.schulz@cherry.de>
Reviewed-by: Kieran Bingham <kieran.bingham@ideasonboard.com>
Signed-off-by: Kieran Bingham <kieran.bingham@ideasonboard.com>
Move black levels for tuning files that contained a BLC block into
the camera sensor helpers.
ov4689.yaml had 66@12bit while the datasheet states 64@12bit. Use the
value from the datasheet (scaled to 16bit).
ov5640.yaml had 256@12bit while the datasheet states 16@10bit. Looking
at the commit message the 256 most likely stems from the imx219 tuning
file and 16@10bit is the same as the 64@12bit from the ov4689. This
seems more likely and is therefore used.
Signed-off-by: Stefan Klug <stefan.klug@ideasonboard.com>
Reviewed-by: Kieran Bingham <kieran.bingham@ideasonboard.com>
Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
For a proper tuning process we need to know the sensor black levels. In
most cases these are fixed and not reported by the kernel driver. Store
them inside the sensor helpers for later retrieval by the algorithms.
Add black level value corresponding to the data pedestal for three
initial sensors as documented in the datasheets. More should be added,
eventually filling the gaps for all supported sensors.
Signed-off-by: Stefan Klug <stefan.klug@ideasonboard.com>
Reviewed-by: Paul Elder <paul.elder@ideasonboard.com>
Reviewed-by: Kieran Bingham <kieran.bingham@ideasonboard.com>
Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
The matrixVlidateYaml() function is declared in the libcamera::ipa::
namespace, but defined in the libcamera:: namespace. This causes a
dynamic linking error at runtime. Fix it by moving the function
definition.
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Tested-by: Jacopo Mondi <jacopo.mondi@ideasonboard.com>
Reviewed-by: Paul Elder <paul.elder@ideasonboard.com>
Add a class to encapsulate the functionality of fetching a matrix based
on an integer key, and interpolating if there is no exact match. This is
expected to be used by both color correction matrices / crosstalk
correction as well as lens shading correction.
A cache is included only for exact matches of the key. The caller is
expected to decide the tolererance for rounding.
Signed-off-by: Paul Elder <paul.elder@ideasonboard.com>
Reviewed-by: Stefan Klug <stefan.klug@ideasonboard.com>
Reviewed-by: Kieran Bingham <kieran.bingham@ideasonboard.com>
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
The Pwl::Pwl(const std::vector<Point> &) constructor is inefficient as
it makes a copy of the given points vector. Add a second constructor
that takes an rvalue reference to a points vector to provide move
semantics.
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Paul Elder <paul.elder@ideasonboard.com>
Reviewed-by: Kieran Bingham <kieran.bingham@ideasonboard.com>
Add a size() function to the Pwl class to return the number of points in
the piecewise linear function. This is useful, for instance, to validate
that all points added with append() or prepend() have been taken into
account.
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Paul Elder <paul.elder@ideasonboard.com>
Reviewed-by: Kieran Bingham <kieran.bingham@ideasonboard.com>
The Pwl::empty() function is a one-liner that can be easily optimized by
the compiler given the chance. Make it inline.
While at it, move the function documentation block to match the class
declaration order.
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Paul Elder <paul.elder@ideasonboard.com>
Reviewed-by: Kieran Bingham <kieran.bingham@ideasonboard.com>
Implement a specialization of the YamlObject::Getter structure to
support deserializing ipa::Vector objects from YAML data.
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Paul Elder <paul.elder@ideasonboard.com>
All CameraSensorHelper subclasses define their member functions inline,
except for the CameraSensorHelperAr0521 class. Inline the gainCode() and
gain() functions to match the other classes.
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Kieran Bingham <kieran.bingham@ideasonboard.com>
Add a vector class to libipa. The original purpose of this is to replace
the floating-point Point class that Raspberry Pi used in their Pwl, as
that implementation of Point seemed more akin to a Vector than a Point.
This is added to libipa instead of to geometry.h to avoid public API
issues, plus this is not expected to be needed by applications.
Signed-off-by: Paul Elder <paul.elder@ideasonboard.com>
Reviewed-by: Stefan Klug <stefan.klug@ideasonboard.com>
Reviewed-by: Kieran Bingham <kieran.bingham@ideasonboard.com>
Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Commit 1dc01bc9e6 ("ipa: libipa: histogram: Add transform parameter to
constructor") added a new constructor for the Histogram class. The
implementation is inline in the header file, and the documentation was
added in the corresponding .cpp file, but it missed the \fn directive.
This causes Doxygen to complain:
src/ipa/libipa/histogram.h:31: warning: Member Histogram(Span< const uint32_t > data, Transform transform) (function) of class libcamera::ipa::Histogram is not documented.
src/ipa/libipa/histogram.cpp:59: warning: libcamera::ipa::Histogram::bins has @param documentation sections but no arguments
Fix it.
Fixes: 1dc01bc9e6 ("ipa: libipa: histogram: Add transform parameter to constructor")
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Umang Jain <umang.jain@ideasonboard.com>
Reviewed-by: Stefan Klug <stefan.klug@ideasonboard.com>
Add a parameter to the histogram constructor that takes a transformation
function to apply to all the bins upon construction.
This is necessary notably for the rkisp1, as the values reported from
the hardware are 20 bits where the upper 16-bits are meaningful integer
values and the lower 4 bits are fractional and meant to be discarded. As
adding a right-shift parameter is probably too specialized, a generic
function is added as a parameter instead.
While at it, optimize the existing constructor to avoid push_back() into
a vector with a known final size.
Signed-off-by: Paul Elder <paul.elder@ideasonboard.com>
Reviewed-by: Kieran Bingham <kieran.bingham@ideasonboard.com>
Reviewed-by: Daniel Scally <dan.scally@ideasonboard.com>
Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Source files in libcamera start by a comment block header, which
includes the file name and a one-line description of the file contents.
While the latter is useful to get a quick overview of the file contents
at a glance, the former is mostly a source of inconvenience. The name in
the comments can easily get out of sync with the file name when files
are renamed, and copy & paste during development have often lead to
incorrect names being used to start with.
Readers of the source code are expected to know which file they're
looking it. Drop the file name from the header comment block.
The change was generated with the following script:
----------------------------------------
dirs="include/libcamera src test utils"
declare -rA patterns=(
['c']=' \* '
['cpp']=' \* '
['h']=' \* '
['py']='# '
['sh']='# '
)
for ext in ${!patterns[@]} ; do
files=$(for dir in $dirs ; do find $dir -name "*.${ext}" ; done)
pattern=${patterns[${ext}]}
for file in $files ; do
name=$(basename ${file})
sed -i "s/^\(${pattern}\)${name} - /\1/" "$file"
done
done
----------------------------------------
This misses several files that are out of sync with the comment block
header. Those will be addressed separately and manually.
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Daniel Scally <dan.scally@ideasonboard.com>
The Agc algorithms for the RkIsp1 and IPU3 IPAs do the same thing in
very large part; following the Rpi IPA's algorithm in spirit with a
few tunable values in that IPA being hardcoded in the libipa ones.
Add a new base class for AgcMeanLuminance which implements the same
algorithm and additionally parses yaml tuning files to inform an IPA
module's Agc algorithm about valid constraint and exposure modes and
their associated bounds.
Reviewed-by: Jacopo Mondi <jacopo.mondi@ideasonboard.com>
Reviewed-by: Stefan Klug <stefan.klug@ideasonboard.com>
Reviewed-by: Paul Elder <paul.elder@ideasonboard.com>
Signed-off-by: Daniel Scally <dan.scally@ideasonboard.com>
Signed-off-by: Kieran Bingham <kieran.bingham@ideasonboard.com>
Laurent Pinchart
Geoffrey Van Landeghem
Jacopo Mondi
Daniel Scally
Stefan Klug
André Apitzsch
Kieran Bingham
Milan Zamazal
Daniel Semkowicz
Paul Elder