Note

This is the documentation for the latest development branch and may refer to features that are not available in released versions. If you are looking for the documentation for a specific release, use the drop-down menu on the left and select the desired version.

NONAI2D CSC API Manual

Overview

The K230 chip has 24 built-in hardware Color Space Conversion (CSC) channels, which can efficiently perform image color space conversions. This module supports various image format conversions including common formats such as RGB/YUV, making it suitable for video processing, image display, and similar applications.

API Reference

Constructor

Function Initialize a CSC conversion channel. The channel number is automatically allocated by the system.

Syntax

from nonai2d import CSC
csc = CSC(fmt, max_width=1920, max_height=1080, buf_num=2)

Parameter Description

Parameter	Type	Description	Default
fmt	int	Target image format (see constant definitions)	Required
max_width	int	Maximum image width supported (in pixels)	1920
max_height	int	Maximum image height supported (in pixels)	1080
buf_num	int	Number of VB buffers allocated, affects processing performance	2

Return Value Returns a CSC object on success, raises an exception on failure.

Important Notes

1. The channel number is automatically allocated by the system (range 0-23); users do not need to and cannot specify it

2. It is recommended to set max_width/max_height according to actual image size to save memory

3. When fmt=PIXEL_FORMAT_GRAYSCALE, the internal pipeline converts to YUV420SP first and then extracts the grayscale channel

Properties

Property	Type	Description
chn	int	Auto-allocated CSC channel number
dst_fmt	int	Target image format

convert Method

Function Perform image color space conversion

Syntax

result = csc.convert(frame, timeout_ms=1000, cvt=True)

Parameter Description

Parameter	Type	Description	Default
frame	py_video_frame_info	Input image frame (from Sensor.snapshot)	Required
timeout_ms	int	Conversion timeout in milliseconds	1000
cvt	bool	True returns an Image object, False returns py_video_frame_info	True

Return Value Returns a converted Image object or py_video_frame_info, depending on the cvt parameter.

get_frame Method

Function Get a converted frame from the CSC channel (low-level interface, does not automatically convert to Image).

Syntax

frame = csc.get_frame(timeout_ms=1000)

Parameter Description

Parameter	Type	Description	Default
timeout_ms	int	Timeout in milliseconds	1000

Return Value Returns a py_video_frame_info object on success, None on timeout.

Note: Frames obtained via get_frame must be manually released using release_frame.

release_frame Method

Function Release a frame obtained via get_frame.

Syntax

csc.release_frame(frame)

Parameter Description

Parameter	Type	Description
frame	py_video_frame_info	The frame object to release

destroy Method

Function Release CSC channel resources

Syntax

csc.destroy()

Description After calling this method, all resources used by the CSC channel will be released, and the object can no longer be used.

Constant Definitions

Image Format Constants

Constant Name	Description	Typical Use Case
PIXEL_FORMAT_GRAYSCALE	Grayscale image format	Black-and-white image processing
PIXEL_FORMAT_RGB_565	RGB565 format (big-endian)	LCD display
PIXEL_FORMAT_RGB_565_LE	RGB565 format (little-endian)	Special display devices
PIXEL_FORMAT_BGR_565	BGR565 format (big-endian)	OpenCV compatibility
PIXEL_FORMAT_BGR_565_LE	BGR565 format (little-endian)	Special display devices
PIXEL_FORMAT_RGB_888	RGB888 format	High-quality image processing
PIXEL_FORMAT_BGR_888	BGR888 format	OpenCV compatibility
PIXEL_FORMAT_ARGB_8888	ARGB8888 format (with transparency)	GUI composition
PIXEL_FORMAT_ARGB_1555	ARGB1555 format (with transparency)	Low color-depth UI
PIXEL_FORMAT_ARGB_4444	ARGB4444 format (with transparency)	Low color-depth UI
PIXEL_FORMAT_BGR_888_PLANAR	BGR888 planar format	Professional image processing
PIXEL_FORMAT_RGB_888_PLANAR	RGB888 planar format	Professional image processing
PIXEL_FORMAT_YVU_PLANAR_420	YVU420 planar format	Video decoding output
PIXEL_FORMAT_YUV_SEMIPLANAR_420	YUV420 semi-planar format	Video encoding input
PIXEL_FORMAT_YVU_SEMIPLANAR_420	YVU420 semi-planar format	Video processing
PIXEL_FORMAT_YVU_PLANAR_444	YVU444 planar format	High-quality video processing
PIXEL_FORMAT_YUV_PACKAGE_444	YUV444 packed format	High-quality video processing
PIXEL_FORMAT_YUV_SEMIPLANAR_444	YUV444 semi-planar format	High-quality video processing
PIXEL_FORMAT_YVU_SEMIPLANAR_444	YVU444 semi-planar format	High-quality video processing

Best Practice Example

import time, os, urandom, sys

from media.display import *
from media.media import *
from media.uvc import *

from nonai2d import CSC

DISPLAY_WIDTH = ALIGN_UP(800, 16)
DISPLAY_HEIGHT = 480

# hardware Color Space Converter (channel auto-allocated)
csc = CSC(CSC.PIXEL_FORMAT_RGB_565)

# use lcd as display output
Display.init(Display.ST7701, width = DISPLAY_WIDTH, height = DISPLAY_HEIGHT, to_ide = True)

while True:
    plugin, dev = UVC.probe()
    if plugin:
        print(f"detect USB Camera {dev}")
        break
    time.sleep_ms(100)

mode = UVC.video_mode(640, 480, UVC.FORMAT_MJPEG, 30)

succ, mode = UVC.select_video_mode(mode)
print(f"select mode success: {succ}, mode: {mode}")

UVC.start(cvt = True)

clock = time.clock()

while True:
    clock.tick()

    img = UVC.snapshot()
    if img is not None:
        img = csc.convert(img)
        Display.show_image(img)

    print(clock.fps())

# deinit display
Display.deinit()
csc.destroy()
UVC.stop()

time.sleep_ms(100)