linear_regression_robust#
# Robust Linear Regression Example
#
# This example shows off how to use the get_regression() method on your CanMV Cam
# to get the linear regression of a ROI. Using this method you can easily build
# a robot which can track lines which all point in the same general direction
# but are not actually connected. Use find_blobs() on lines that are nicely
# connected for better filtering options and control.
#
# We're using the robust=True argument for get_regression() in this script which
# computes the linear regression using a much more robust algorithm... but potentially
# much slower. The robust algorithm runs in O(N^2) time on the image. So, YOU NEED
# TO LIMIT THE NUMBER OF PIXELS the robust algorithm works on or it can actually
# take seconds for the algorithm to give you a result... THRESHOLD VERY CAREFULLY!
from media.camera import *
from media.display import *
from media.media import *
import time, os, gc, sys
DISPLAY_WIDTH = ALIGN_UP(1920, 16)
DISPLAY_HEIGHT = 1080
SCALE = 4
DETECT_WIDTH = DISPLAY_WIDTH // SCALE
DETECT_HEIGHT = DISPLAY_HEIGHT // SCALE
THRESHOLD = (0, 100) # Grayscale threshold for dark things...
BINARY_VISIBLE = True # Does binary first so you can see what the linear regression
# is being run on... might lower FPS though.
def camera_init():
# use hdmi for display
display.init(LT9611_1920X1080_30FPS)
# config vb for osd layer
config = k_vb_config()
config.max_pool_cnt = 1
config.comm_pool[0].blk_size = 4*DISPLAY_WIDTH*DISPLAY_HEIGHT
config.comm_pool[0].blk_cnt = 1
config.comm_pool[0].mode = VB_REMAP_MODE_NOCACHE
# meida buffer config
media.buffer_config(config)
# init default sensor
camera.sensor_init(CAM_DEV_ID_0, CAM_DEFAULT_SENSOR)
# set chn0 output size
camera.set_outsize(CAM_DEV_ID_0, CAM_CHN_ID_0, DISPLAY_WIDTH, DISPLAY_HEIGHT)
# set chn0 output format
camera.set_outfmt(CAM_DEV_ID_0, CAM_CHN_ID_0, PIXEL_FORMAT_YUV_SEMIPLANAR_420)
# create meida source device
globals()["meida_source"] = media_device(CAMERA_MOD_ID, CAM_DEV_ID_0, CAM_CHN_ID_0)
# create meida sink device
globals()["meida_sink"] = media_device(DISPLAY_MOD_ID, DISPLAY_DEV_ID, DISPLAY_CHN_VIDEO1)
# create meida link
media.create_link(meida_source, meida_sink)
# set display plane with video channel
display.set_plane(0, 0, DISPLAY_WIDTH, DISPLAY_HEIGHT, PIXEL_FORMAT_YVU_PLANAR_420, DISPLAY_MIRROR_NONE, DISPLAY_CHN_VIDEO1)
# set chn1 output nv12
camera.set_outsize(CAM_DEV_ID_0, CAM_CHN_ID_1, DETECT_WIDTH, DETECT_HEIGHT)
camera.set_outfmt(CAM_DEV_ID_0, CAM_CHN_ID_1, PIXEL_FORMAT_YUV_SEMIPLANAR_420)
# media buffer init
media.buffer_init()
# request media buffer for osd image
globals()["buffer"] = media.request_buffer(4 * DISPLAY_WIDTH * DISPLAY_HEIGHT)
# start stream for camera device0
camera.start_stream(CAM_DEV_ID_0)
def camera_deinit():
# stop stream for camera device0
camera.stop_stream(CAM_DEV_ID_0)
# deinit display
display.deinit()
os.exitpoint(os.EXITPOINT_ENABLE_SLEEP)
time.sleep_ms(100)
# release media buffer
media.release_buffer(globals()["buffer"])
# destroy media link
media.destroy_link(globals()["meida_source"], globals()["meida_sink"])
# deinit media buffer
media.buffer_deinit()
def capture_picture():
# create image for osd
buffer = globals()["buffer"]
osd_img = image.Image(DETECT_WIDTH, DETECT_HEIGHT, image.GRAYSCALE, alloc=image.ALLOC_VB, phyaddr=buffer.phys_addr, virtaddr=buffer.virt_addr, poolid=buffer.pool_id)
osd_img.clear()
osd_img.draw_string(0, 0, "Please wait...")
display.show_image(osd_img, 0, 0, DISPLAY_CHN_OSD0)
fps = time.clock()
while True:
fps.tick()
try:
os.exitpoint()
yuv420_img = camera.capture_image(CAM_DEV_ID_0, CAM_CHN_ID_1)
img = image.Image(yuv420_img.width(), yuv420_img.height(), image.GRAYSCALE, data=yuv420_img)
camera.release_image(CAM_DEV_ID_0, CAM_CHN_ID_1, yuv420_img)
img = img.binary([THRESHOLD]) if BINARY_VISIBLE else img
# Returns a line object similar to line objects returned by find_lines() and
# find_line_segments(). You have x1(), y1(), x2(), y2(), length(),
# theta() (rotation in degrees), rho(), and magnitude().
#
# magnitude() represents how well the linear regression worked. It goes from
# (0, INF] where 0 is returned for a circle. The more linear the
# scene is the higher the magnitude.
line = img.get_regression([(255,255) if BINARY_VISIBLE else THRESHOLD], robust = True)
if (line): img.draw_line(line.line(), color = 127)
print("FPS %f, mag = %s" % (fps.fps(), str(line.magnitude()) if (line) else "N/A"))
img.copy_to(osd_img)
del img
gc.collect()
except KeyboardInterrupt as e:
print("user stop: ", e)
break
except BaseException as e:
sys.print_exception(e)
break
def main():
os.exitpoint(os.EXITPOINT_ENABLE)
camera_is_init = False
try:
print("camera init")
camera_init()
camera_is_init = True
print("camera capture")
capture_picture()
except Exception as e:
sys.print_exception(e)
finally:
if camera_is_init:
print("camera deinit")
camera_deinit()
if __name__ == "__main__":
main()
具体接口使用请参考相关文档说明: