期待结果和实际结果
我希望一个sensor2运行AI程序、sensor1运行检测色块之类的程序
但尝试了很多次一直报各种错或者效果不理想、有没有大佬能帮助一下
重现步骤
以下是我的双摄像头显示程序
#作 用： 导入模块
from media.sensor import *
from media.display import *
from media.media import *
from libs.PlatTasks import DetectionApp
from libs.Utils import *
import image
import time, gc, ujson
#------------------------------------定义变量区---------------------------------------------------
sensor2 = None
sensor1=None
#-----------------------------------------------------------------------------------------------
#作 用： 主函数执行
#填入值：无
#返回值：无
def main():
try:
#------------------------------------构建摄像头与显示对象-----------------------------------------
#构建 Sensor 对象 sensor2
sensor2 = Sensor(id=2)
sensor2.reset()
sensor2.set_framesize(width = 400, height = 240, chn = CAM_CHN_ID_0)
sensor2.set_pixformat(Sensor.RGB565, chn = CAM_CHN_ID_0)
sensor2.set_framesize(width=400,height=240, chn = CAM_CHN_ID_1)
sensor2.set_pixformat(Sensor.RGB888, chn = CAM_CHN_ID_1)
# 构建 Sensor 对象 sensor1
sensor1 = Sensor(id=1)
sensor1.reset()
sensor1.set_framesize(width=400, height=240, chn = CAM_CHN_ID_0)
sensor1.set_pixformat(Sensor.RGB565, chn = CAM_CHN_ID_0)
sensor1.set_framesize(width=400,height=240,chn = CAM_CHN_ID_1)
sensor1.set_pixformat(Sensor.RGB565, chn = CAM_CHN_ID_1)
sensor1.set_hmirror(True)
sensor1.set_vflip(True)
#构建Display对象
Display.init(Display.ST7701, to_ide = True, osd_num = 2)
MediaManager.init()
sensor1.run()
sensor2.run()
#----------------------------------------------------------------------------------------------
clock = time.clock() #初始化帧率
while True:
clock.tick() #记录当前时间（毫秒），可用于FPS计算
os.exitpoint()
img1 = sensor1.snapshot(chn = CAM_CHN_ID_0)
img2 = sensor2.snapshot(chn = CAM_CHN_ID_0)
#------------------------------------主要函数区--------------------------------------------------
#----------------------------------------------------------------------------------------------
fps =f'FPS: {clock.fps():.3f}' #接收帧率信息
img1.draw_string_advanced(0,0,20,fps,color=(255,255,255)) #绘制出帧率信息
Display.show_image(img1, x=0,y=0, layer = Display.LAYER_OSD0)
Display.show_image(img2, x=400,y=240, layer = Display.LAYER_OSD1)
except KeyboardInterrupt as e:
print("user stop: ", e)
except BaseException as e:
print(f"Exception {e}")
finally:
os.exitpoint(os.EXITPOINT_ENABLE_SLEEP)
sensor1.stop()
sensor2.stop()
# deinit display
Display.deinit()
time.sleep_ms(100)
# release media buffer
MediaManager.deinit()
#作 用： 设置主函数
#填入值：无
#返回值：无
if __name__=="__main__":
main()
以下是我的AI程序
# -*- coding: utf-8 -*-
'''
Script: deploy_det_video.py
脚本名称：deploy_det_video.py
Description:
This script runs a real-time object detection application on an embedded device.
It uses a pipeline to capture video frames, performs inference using a pre-trained Kmodel,
and displays the detection results (bounding boxes, class labels) on screen.
The model configuration is loaded from the Canaan online training platform via a JSON config file.
脚本说明：
本脚本在嵌入式设备上运行实时目标检测应用。它通过捕获视频帧，使用预训练的 Kmodel 进行推理，并在屏幕上显示检测结果（边界框、类别标签）。
模型配置文件通过 Canaan 在线训练平台从 JSON 文件加载。
Author: Canaan Developer
作者：Canaan 开发者
'''
import os, gc
from libs.PlatTasks import DetectionApp
from libs.PipeLine import PipeLine
from libs.Utils import *
# Set display mode: options are 'hdmi', 'lcd', 'lt9611', 'st7701', 'hx8399'
# 'hdmi' defaults to 'lt9611' (1920x1080); 'lcd' defaults to 'st7701' (800x480)
display_mode = "lt9611"
# Define the input size for the RGB888P video frames
rgb888p_size = [1280, 720]
# Set root directory path for model and config
root_path = "/sdcard/mp_deployment_source/"
# Load deployment configuration
deploy_conf = read_json(root_path + "/deploy_config.json")
kmodel_path = root_path + deploy_conf["kmodel_path"] # KModel path
labels = deploy_conf["categories"] # Label list
confidence_threshold = deploy_conf["confidence_threshold"] # Confidence threshold
nms_threshold = deploy_conf["nms_threshold"] # NMS threshold
model_input_size = deploy_conf["img_size"] # Model input size
nms_option = deploy_conf["nms_option"] # NMS strategy
model_type = deploy_conf["model_type"] # Detection model type
anchors = []
if model_type == "AnchorBaseDet":
anchors = deploy_conf["anchors"][0] + deploy_conf["anchors"][1] + deploy_conf["anchors"][2]
# Inference configuration
inference_mode = "video" # Inference mode: 'video'
debug_mode = 0 # Debug mode flag
# Create and initialize the video/display pipeline
pl = PipeLine(rgb888p_size=rgb888p_size, display_mode=display_mode)
pl.create()
display_size = pl.get_display_size()
# Initialize object detection application
det_app = DetectionApp(inference_mode,kmodel_path,labels,model_input_size,anchors,model_type,confidence_threshold,nms_threshold,rgb888p_size,display_size,debug_mode=debug_mode)
# Configure preprocessing for the model
det_app.config_preprocess()
# Main loop: capture, run inference, display results
while True:
with ScopedTiming("total", 1):
img = pl.get_frame() # Capture current frame
res = det_app.run(img) # Run inference
det_app.draw_result(pl.osd_img, res) # Draw detection results
pl.show_image() # Show result on display
gc.collect() # Run garbage collection
# Cleanup: These lines will only run if the loop is interrupted (e.g., by an IDE break or external interruption)
det_app.deinit() # De-initialize detection app
pl.destroy() # Destroy pipeline instance
以下是我的色块检测类程序
import time, os, sys
from machine import UART
from machine import FPIOA
from media.sensor import *
from media.display import *
from media.media import *
try:
# 配置引脚
fpioa = FPIOA()
fpioa.set_function(11, FPIOA.UART2_TXD)
fpioa.set_function(12, FPIOA.UART2_RXD)
# 初始化UART2，波特率115200，8位数据位，无校验，1位停止位
uart = UART(UART.UART2, baudrate=115200, bits=UART.EIGHTBITS, parity=UART.PARITY_NONE, stop=UART.STOPBITS_ONE)
# 获取传感器数据（假设为变量sensor_value）
sensor_value = 0 # 示例数据
# 构造一个具有默认配置的摄像头对象
sensor = Sensor(id=2)
# 重置摄像头sensor
sensor.reset()
# 设置水平镜像和设置垂直翻转
# sensor.set_hmirror(False) sensor.set_vflip(False)
# 设置通道0的输出尺寸为显示分辨率
sensor.set_framesize(width=800, height=480, chn=CAM_CHN_ID_0)
# 设置通道0的输出像素格式为RGB565
sensor.set_pixformat(Sensor.RGB565, chn=CAM_CHN_ID_0)
# 根据模式初始化显示器
Display.init(Display.ST7701, width=800, height=480, to_ide=True)
# 初始化媒体管理器
MediaManager.init()
# 启动传感器
sensor.run()
while True:
# 捕获通道0的图像
img = sensor.snapshot(chn=CAM_CHN_ID_0)
#color_threshold 是要寻找的颜色的阈值，area_threshold 表示过滤掉小于此面积的色块。
blobs = img.find_blobs([(34, 92, -49, 127, -61, 50)],roi=(0,240,800,240),area_threshold = 10000)
if blobs:
# 遍历每个检测到的颜色块
for blob in blobs:
# 绘制颜色块的外接矩形
# blob[0:4] 表示颜色块的矩形框 [x, y, w, h]，
img.draw_rectangle(blob[0:4])
# 在颜色块的中心绘制一个十字
# blob[5] 和 blob[6] 分别是颜色块的中心坐标 (cx, cy)
img.draw_line(blob[5],240,blob[5],480,color=(0,255,255),thickness=5)
# 中线
img.draw_line(400,0,400,480,color=(0,0,0),thickness=3)
# 在控制台输出颜色块的中心坐标
print("Blob Center: X={}, Y={}".format(blob[5], blob[6]))
message = "Y:{}\n".format(blob[5])
uart.write(message)
# 显示捕获的图像，中心对齐，居中显示
Display.show_image(img)
os.exitpoint()
except KeyboardInterrupt as e:
print("用户停止: ", e)
except BaseException as e:
print(f"异常: {e}")
finally:
# 停止传感器运行
if isinstance(sensor, Sensor):
sensor.stop()
# 反初始化显示模块
Display.deinit()
os.exitpoint(os.EXITPOINT_ENABLE_SLEEP)
time.sleep_ms(100)
# 释放媒体缓冲区
MediaManager.deinit()
软硬件版本信息
庐山派+屏幕+最新固件+canMV IDE
补充材料
