# -*- coding:utf-8 -*- # @Author len # @Create 2023/11/23 11:28 import math import sensor, image, time, lcd import binascii from Maix import GPIO from machine import Timer, PWM, UART, Timer from fpioa_manager import fm import KPU as kpu class Mainlen(): def __init__(self): '''初始化摄像头和 LCD 显示屏''' lcd.init() # lcd初始化 self.canera_init() # 摄像头初始化 # sensor.set_auto_gain(0, gain_db=17) # 设置摄像头的自动增益功能 # 映射串口引脚 fm.register(6, fm.fpioa.UART1_RX, force=True) fm.register(7, fm.fpioa.UART1_TX, force=True) # 初始化串口 self.uart = UART(UART.UART1, 115200, read_buf_len=4096) # 循迹 # --------------感光芯片配置 START ------------------- self.IMG_WIDTH = 240 self.IMG_HEIGHT = 320 # 直线灰度图颜色阈值 self.LINE_COLOR_THRESHOLD = [(0, 60)] # 找黑色 self.LINE_COLOR_BAISE = [(60, 255)] # 找白色 self.ROIS = { # 找黑色 # 'left': (0, 0, 320, 50), # 纵向取样-左侧 # 'right': (0, 190, 320, 50), # 纵向取样-右侧 'left': (0, 0, 180, 50), # 纵向取样-左侧 'right': (0, 190, 180, 50), # 纵向取样-右侧 'up': (240, 0, 80, 240), # 横向取样-上方 'middle_up': (160, 0, 80, 240), # 横向取样-中上 'middle_down': (80, 0, 80, 240), # 横向取样-中下 'down': (0, 0, 80, 240), # 横向取样-下方 } # 红黄绿的阈值 # 颜色识别阈值 (L Min, L Max, A Min, A Max, B Min, B Max) LAB模型 # 下列阈值元组用来识别 红、绿、蓝三种颜色，可通过调整数据阈值完成更多颜色的识别。 # self.thresholds = [(59, 100, 40, 127, 5, 127), # 红色阈值 # (90, 100, -5, 2, -4, 20), # 黄色阈值 # (87, 100, -59, 127, -10, 127),] # 绿色阈值 self.thresholds = [(56, 99, 13, 63, -18, 37, "红色"), # 红色阈值 (69, 100, -7, 14, 1, 46, "黄色"), # 黄色阈值 (55, 100, -51, -16, -9, 83, "绿色")] # 绿色阈值 self.is_need_send_data = False # 是否需要发送数据的信号标志 self.QRFlag = 0 # 主函数 run def startMain(self): Flag_track = False # 循迹标识 Flag_qr = False # 二维码标识 Flag_qr_2 = False # 标识B Flag_qr_3 = False # 二维码3 Flag_light = False # 交通灯标识 Flag_Camera_wheel = False # 摄像头转弯 while True: data = self.uart.read(8) # print(data) # print(len(data)) # if (len(data) >= 8): if data is not None: if (data[1] == 0x02) and (data[7] == 0xBB) and self.verify_checksum(data): # 巡线与控制舵机 if data[2] == 0x91: if data[3] == 0x01: # 启动循迹 sensor.set_pixformat(sensor.GRAYSCALE) # 设置像素格式为灰色 Flag_track = True print("开始循迹") elif data[3] == 0x02: # 停止循迹 sensor.set_pixformat(sensor.RGB565) # 设置像素格式为彩色 RGB565 Flag_track = False print("停止循迹") elif data[3] == 0x03: # 调整舵机 print("调整舵机") self.Servo(data) elif data[3] == 0x04: # 启动摄像头转弯 print("摄像头转弯") Flag_Camera_wheel = True elif data[3] == 0x05: # 停止摄像头转弯 print("停止摄像头转弯") Flag_Camera_wheel = False else: pass # 识别任务 elif data[2] == 0x92: if data[3] == 0x01: # 识别二维码 sensor.set_pixformat(sensor.RGB565) # 设置像素格式为彩色 RGB565 Flag_qr = True print("开始识别二维码") elif data[3] == 0x02: # 停止识别二维码 Flag_qr = False print("停止识别二维码") elif data[3] == 0x03: # 识别交通灯 sensor.set_pixformat(sensor.RGB565) # 设置像素格式为彩色 RGB565 Flag_light = True print("开始识别交通灯") elif data[3] == 0x04: # 停止识别交通灯 Flag_light = False print("停止识别交通灯") elif data[3] == 0x06: # 调整摄像头阈值 print("调整摄像头阈值") self.canera_ash() elif data[3] == 0x09: print("开始识别第二个二维码") Flag_qr_2 = True elif data[3] == 0x10: print("停止识别第二个二维码") Flag_qr_2 = False elif data[3] == 0x12: print("开始识别第三个二维码") Flag_qr_3 = True elif data[3] == 0x13: print("停止识别第三个二维码") Flag_qr_3 = False else: pass img = sensor.snapshot() # 获取图像 if Flag_track: # 循迹 print("循迹") self.tracking(img) elif Flag_Camera_wheel: # 摄像头转弯 print("摄像头转弯") self.Camera_wheel(img) elif Flag_qr and (self.QRFlag <3): # 二维码 print("识别二维码") self.discem_QR(img) Flag_qr = False elif Flag_light: # 红绿灯 print("识别红绿灯") self.discem_light() # Flag_light = False elif Flag_qr_2: print("识别第二个二维码") self.discem_QR_2(img) Flag_qr_2 = False elif Flag_qr_3: print("识别第三个二维码") self.discem_QR_3(img) Flag_qr_3 = False lcd.display(img) # 在LCD显示 # 摄像头转弯 def Camera_wheel(self, img): canvas = img.copy() roi_blobs_result = {} # 在各个ROI中寻找色块的结果记录 for roi_direct in self.ROIS.keys(): roi_blobs_result[roi_direct] = { 'cx': 0, 'cy': 0, 'w': 0, 'blob_flag': False } for roi_direct, roi in self.ROIS.items(): blobs = canvas.find_blobs(self.LINE_COLOR_THRESHOLD, roi=roi, merge=True) if len(blobs) != 0: largest_blob = max(blobs, key=lambda b: b.pixels()) if largest_blob.area() > 1000: roi_blobs_result[roi_direct]['cx'] = largest_blob.cy() roi_blobs_result[roi_direct]['cy'] = largest_blob.cx() roi_blobs_result[roi_direct]['w'] = largest_blob.h() roi_blobs_result[roi_direct]['blob_flag'] = True x, y, width, height = largest_blob[:4] img.draw_rectangle((x, y, width, height), color=(255)) else: # blobs=canvas.find_blobs(LINE_COLOR_THRESHOLD, roi=roi, merge=True, pixels_area=10) continue blobs_zuo = canvas.find_blobs(self.LINE_COLOR_THRESHOLD, roi=(0, 0, 180, 50), merge=True) # 车载摄像头屏幕左下部找黑色 blobs_you = canvas.find_blobs(self.LINE_COLOR_THRESHOLD, roi=(0, 190, 180, 50), merge=True) # 车载摄像头屏幕右下部找黑色 print(blobs_zuo) print(blobs_you) if len(blobs_zuo) != 0: zuo = blobs_zuo[0][4] else: zuo = 0 if len(blobs_you) != 0: you = blobs_you[0][4] else: you = 0 print(zuo, you) if zuo >= 8000 or you >= 8000: print("黑线") wire = 0x01 else: print("白线") wire = 0x00 # print() self.uart.write(bytes([0x55])) self.uart.write(bytes([0x02])) self.uart.write(bytes([0x93])) self.uart.write(bytes([0x03])) # 线 self.uart.write(bytes([wire])) # 黑白线 self.uart.write(bytes([0x03])) self.uart.write(bytes([0x03])) self.uart.write(bytes([0xbb])) # 初始化摄像头阈值 def canera_init(self): # 摄像头模块初始化 sensor.reset() # 复位和初始化摄像头 sensor.set_pixformat(sensor.RGB565) # 设置像素格式为彩色 RGB565 # sensor.set_pixformat(sensor.GRAYSCALE) # 设置像素格式为灰色 sensor.set_framesize(sensor.QVGA) # 设置帧大小为QVGA（320×240） sensor.set_vflip(1) # 后置模式 sensor.skip_frames(30) # # 跳过前30帧 # 调整摄像头阈值 def canera_ash(self): sensor.reset() # 复位和初始化摄像头 sensor.set_vflip(1) # 将摄像头设置成后置方式（所见即所得） sensor.set_pixformat(sensor.RGB565) # 设置像素格式为彩色 RGB565 # sensor.set_pixformat(sensor.GRAYSCALE) # 设置像素格式为灰色 sensor.set_framesize(sensor.QVGA) # 设置帧大小为 QVGA (320x240) # sensor.set_windowing((224, 224)) # 设置摄像头的窗口大小 sensor.set_auto_gain(False) sensor.set_auto_whitebal(False) sensor.set_auto_gain(0, 0) sensor.skip_frames(time=200) # 等待设置生效 # 红绿灯 def discem_light(self): # 初始化 results = [] start_time = time.time() while time.time() - start_time < 3: # 循环持续三秒 img = sensor.snapshot() img = img.crop((160, 0, 80, 224)) # print(img.width()) # 320 # print(img.height()) #240 # img = img.crop((0, 0, 160, 224)) max_blob = None max_blob_size = 0 max_blob_color = "" max_blob_color_index = 0 # 找圆 # circles = img.find_circles(threshold=3500, x_margin=10, y_margin=10, r_margin=10, # r_min=2, r_max=100, r_step=2) for index, threshold in enumerate(self.thresholds): # 查找每种颜色的色块 blobs = img.find_blobs([threshold[:-1]], pixels_threshold=150, area_threshold=150, merge=True) if blobs: # 找到最大的色块 blob = max(blobs, key=lambda b: b.pixels()) if blob.pixels() > max_blob_size: max_blob = blob max_blob_size = blob.pixels() max_blob_color = threshold[-1] # 颜色标签 max_blob_color_index = index # print(circles) # # 画圆 # for c in circles: # area = (c.x() - c.r(), c.y() - c.r(), 2 * c.r(), 2 * c.r()) # img.draw_rectangle(area, color=(255, 255, 255)) # 绘制最大色块的矩形和中心十字，并输出颜色 if max_blob: results.append((max_blob_color, max_blob_color_index)) # 存储识别结果 img.draw_rectangle(max_blob[0:4]) img.draw_cross(max_blob[5], max_blob[6]) print("最大色块的颜色是:", max_blob_color, max_blob_color_index) # 初始化计数字典 color_count = {} for color, index in results: if (color, index) in color_count: color_count[(color, index)] += 1 else: color_count[(color, index)] = 1 # 找出出现次数最多的颜色和下标 if color_count: most_common_color, most_common_index = max(color_count, key=color_count.get) print("出现最多的颜色:", most_common_color, "在下标位置:", most_common_index) else: most_common_index = 1 print("没有识别到有效的颜色块") # 串口发送交通灯信息 # send_data = [ # 0x55, # 0x02, # 0x92, # 0x03, # most_common_index, # 0xbb] # print("红绿灯指令", send_data, "******************", bytes(send_data)) # self.UsartSend(bytes(send_data)) self.uart.write(bytes([0x55])) self.uart.write(bytes([0x02])) self.uart.write(bytes([0x92])) self.uart.write(bytes([0x03])) # 红绿灯 self.uart.write(bytes([most_common_index])) # 红绿灯结果 self.uart.write(bytes([0x03])) self.uart.write(bytes([0x03])) self.uart.write(bytes([0xbb])) self.canera_init() # 恢复摄像头 # 二维码 def discem_QR(self, img): # self.Tise_servo(10) res_QR = img.find_qrcodes() # 寻找二维码 timeflag = 0 while timeflag < 30: img = sensor.snapshot() # 获取图像 TS_QR = img.find_qrcodes() # 再次寻找二维码 for qr in TS_QR: if all(qr.payload() != existing_qr.payload() for existing_qr in res_QR): res_QR.append(qr) timeflag += 1 time.sleep(0.001) # print(timeflag) print(len(res_QR)) if len(res_QR): # 识别到 for res in range(len(res_QR)): count = 0 if_end = 0x01 result = res_QR[res].payload() print(result) result = self.qr_240606(result) if self.if_formula(result): # 公式 count = 2 else: # 车牌 result = self.extract_numbers_and_letters_combined(result) count = 1 if res == len(res_QR) - 1: if_end = 0x00 # if_end = 0x00 print("count: ", count) print("if_end: ", if_end) # 串口发送二维码信息 self.uart.write(bytes([0x55])) self.uart.write(bytes([0x02])) self.uart.write(bytes([0x92])) self.uart.write(bytes([0x06])) self.uart.write(bytes([count])) self.uart.write(bytes([if_end])) self.uart.write(bytes([len(result)])) for qr_data in result: self.uart.write(bytes([ord(qr_data)])) print(bytes([ord(qr_data)])) self.uart.write(bytes([0xbb])) time.sleep(1) else: # 未识别 self.uart.write(bytes([0x55])) self.uart.write(bytes([0x02])) self.uart.write(bytes([0x92])) self.uart.write(bytes([0x06])) self.uart.write(bytes([0xff])) self.uart.write(bytes([0x00])) self.uart.write(bytes([0x00])) self.uart.write(bytes([0xbb])) # time.sleep(1) def discem_QR_2(self, img): # self.Tise_servo(10) res_QR = img.find_qrcodes() # 寻找二维码 timeflag = 0 while timeflag < 30: img = sensor.snapshot() # 获取图像 TS_QR = img.find_qrcodes() # 再次寻找二维码 for qr in TS_QR: if all(qr.payload() != existing_qr.payload() for existing_qr in res_QR): res_QR.append(qr) timeflag += 1 time.sleep(0.001) # print(timeflag) print(len(res_QR)) if len(res_QR) == 2: qr_code_a = res_QR[0].payload() qr_code_b = res_QR[1].payload() ascii_codes = self.qr_240606_2(qr_code_a, qr_code_b) # 最大子串 print(ascii_codes) # # 计算前6个字符的ASCII码 # ascii_codes = [hex(ord(c)) for c in sorted_combined[:6]] # print(ascii_codes) # 串口发送二维码信息 self.uart.write(bytes([0x55])) self.uart.write(bytes([0x02])) self.uart.write(bytes([0x92])) self.uart.write(bytes([0x07])) self.uart.write(bytes([0x01])) # self.uart.write(bytes([if_end])) # # self.uart.write(bytes([len(result_list)])) for qr_data in ascii_codes: self.uart.write(bytes([ord(qr_data)])) print(bytes([ord(qr_data)])) self.uart.write(bytes([0xbb])) time.sleep(1) # for res in range(len(res_QR)): # count = 0 # if_end = 0x01 # result = res_QR[res].payload() # print(result) # # if len(result) == 8: # 车牌 # result = self.extract_numbers_and_letters_combined(result) # count = 1 # # elif self.if_formula(result): # 公式 # count = 2 # # continue # # if res == len(res_QR) - 1: # if_end = 0x00 # # if_end = 0x00 # print("count: ", count) # print("if_end: ", if_end) else: # 未识别 self.uart.write(bytes([0x55])) self.uart.write(bytes([0x02])) self.uart.write(bytes([0x92])) self.uart.write(bytes([0x07])) self.uart.write(bytes([0xff])) self.uart.write(bytes([0x00])) self.uart.write(bytes([0x00])) self.uart.write(bytes([0xbb])) # time.sleep(1) def discem_QR_3(self, img): # self.Tise_servo(10) res_QR = img.find_qrcodes() # 寻找二维码 timeflag = 0 while timeflag < 30: img = sensor.snapshot() # 获取图像 TS_QR = img.find_qrcodes() # 再次寻找二维码 for qr in TS_QR: if all(qr.payload() != existing_qr.payload() for existing_qr in res_QR): res_QR.append(qr) timeflag += 1 time.sleep(0.001) # print(timeflag) print(len(res_QR)) if len(res_QR) == 2: qr_code_a = res_QR[0].payload() qr_code_b = res_QR[1].payload() ascii_codes = self.qr_240606(qr_code_a, qr_code_b) # 最大子串 print(ascii_codes) # 串口发送二维码信息 self.uart.write(bytes([0x55])) self.uart.write(bytes([0x02])) self.uart.write(bytes([0x92])) self.uart.write(bytes([0x08])) self.uart.write(bytes([0x01])) # self.uart.write(bytes([if_end])) # # self.uart.write(bytes([len(result_list)])) for qr_data in ascii_codes: self.uart.write(bytes([ord(qr_data)])) print(bytes([ord(qr_data)])) self.uart.write(bytes([0xbb])) time.sleep(1) else: # 未识别 self.uart.write(bytes([0x55])) self.uart.write(bytes([0x02])) self.uart.write(bytes([0x92])) self.uart.write(bytes([0x08])) self.uart.write(bytes([0xff])) self.uart.write(bytes([0x00])) self.uart.write(bytes([0x00])) self.uart.write(bytes([0xbb])) # time.sleep(1) def qr_240606(self, str1, str2): result = "" ch1 = self.get_test_outside_brackets(str1) ch2 = self.get_test_outside_brackets(str2) str1 = self.get_test_inside_brackets(str1) str2 = self.get_test_inside_brackets(str2) # print(str1, str2) if ch1 == 'A': result = self.compare_strings(str1, str2) elif ch2 == 'A': result = self.compare_strings(str2, str1) return result # 比较两个字符串的内容，存第一个二维码的不同 def compare_strings(self, str1, str2): result1 = "" result2 = "" for i in range(len(str1)): if str1[i] != str2[i]: result1 += str1[i] result2 += str2[i] return result1 + result2 # 尖括号外的数据 def get_test_outside_brackets(self, qrstring): # 提取尖括号外的内容 start_index = qrstring.find('<') end_index = qrstring.find('>') # 如果尖括号存在，切割尖括号外的内容 if start_index != -1 and end_index != -1: return qrstring[:start_index] + qrstring[end_index + 1:] else: # 如果尖括号不存在，返回原字符串 return qrstring # 尖括号内的数据 def get_test_inside_brackets(self, qrstring): # 提取尖括号内的内容 start_index = qrstring.find('<') end_index = qrstring.find('>') # 如果尖括号存在，切割尖括号内的内容 if start_index != -1 and end_index != -1: return qrstring[start_index + 1:end_index] else: # 如果尖括号不存在，返回空字符串 return "" def qr_240606_2(self, red, green): result_1 = "" result_2 = "" nei1 = self.extract_innermost_bracket_content(red) nei2 = self.extract_innermost_bracket_content(green) one = red.replace(nei1, '') two = green.replace(nei2, '') nei1 = self.extract_numbers_without_isdigit(nei1) nei2 = self.extract_numbers_without_isdigit(nei2) # print(one, two) one_number = self.extract_numbers_without_isdigit(one) two_number = self.extract_numbers_without_isdigit(two) one_str = self.extract_letters_without_isalpha(one) two_str = self.extract_letters_without_isalpha(two) # print(one_number, one_str) # print(two_number, two_str) if len(one_number) == 2 and len(two_str) == 2: result_1 = one_number + nei1 result_2 = two_str + nei2 elif len(one_str) == 2 and len(two_number) == 2: result_1 = two_number + nei2 result_2 = one_number + nei1 return result_1 + result_2 # 提取数字字符 def extract_numbers_without_isdigit(self, text): numbers_str = '' for char in text: # 检查字符是否是数字（0-9） if '0' <= char <= '9': numbers_str += char return numbers_str # 提取字母字符 def extract_letters_without_isalpha(self, text): letters_str = '' for char in text: # 检查字符是否是字母（A-Z 或 a-z） if ('A' <= char <= 'Z') or ('a' <= char <= 'z'): letters_str += char return letters_str def extract_innermost_bracket_content(self, input_string): """ 提取给定字符串中最内层尖括号<>内的内容。 参数: input_string (str): 输入的字符串。 返回: str: 最内层尖括号内的内容，如果没有匹配则返回空字符串。 """ depth = 0 # 嵌套深度计数器 max_depth = 0 # 最大深度 content_at_depth = {} # 存储每个深度的内容 for char in input_string: if char == '<': depth += 1 max_depth = max(max_depth, depth) # 更新最大深度 content_at_depth[depth] = "" # 为新的深度初始化内容 elif char == '>': if depth > 0: # 只有在有效的深度下才减少深度 depth -= 1 elif depth > 0: # 只有在尖括号内部时才添加内容 content_at_depth[depth] += char # 从记录的最大深度获取内容 return content_at_depth.get(max_depth, "") # 提取字符和数字 def extract_numbers_and_letters_combined(self, text): combined_str = '' for char in text: # 检查字符是否是数字（0-9）或字母（A-Z 或 a-z） if ('0' <= char <= '9') or ('A' <= char <= 'Z') or ('a' <= char <= 'z'): combined_str += char return combined_str def find_longest_common_substring(self, s1, s2): # 初始化动态规划表 m, n = len(s1), len(s2) dp = [[0] * (n + 1) for _ in range(m + 1)] longest, x_longest = 0, 0 for i in range(1, m + 1): for j in range(1, n + 1): if s1[i - 1] == s2[j - 1]: dp[i][j] = dp[i - 1][j - 1] + 1 if dp[i][j] > longest: longest = dp[i][j] x_longest = i else: dp[i][j] = 0 # 返回最长公共子字符串 return s1[x_longest - longest: x_longest] def if_formula(self, text): # 允许的字符：数字、运算符和括号 allowed_chars = "0123456789+-*/%^(). abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ" # 检查字符是否合法 for char in text: if char not in allowed_chars: return False # 括号匹配检查 bracket_count = 0 for char in text: if char == '(': bracket_count += 1 elif char == ')': bracket_count -= 1 if bracket_count < 0: # 右括号多于左括号 return False if bracket_count != 0: # 左括号和右括号数量不匹配 return False return True # 移除字符串内的中文 def remove_chinese_chars(self, text): """移除字符串中的中文字符，并返回新的字符串""" # 定义中文字符的Unicode范围 chinese_char_ranges = [ ('\u4e00', '\u9fff'), # 基本汉字 ('\u3400', '\u4dbf'), # 扩展A ('\u20000', '\u2a6df'), # 扩展B # 可以根据需要添加更多的中文范围 ] # 移除中文字符 return ''.join(char for char in text if not any(start <= char <= end for start, end in chinese_char_ranges)) # 循迹 def tracking(self, img): print("循迹") roi_blobs_result = self.find_blobs_in_rois(img) down_center, state_crossing, deflection_angle = self.state_deflection_angle(roi_blobs_result) dsd = self.data_format_wrapper(down_center, state_crossing, deflection_angle) self.UsartSend(dsd) print("下发指令：", dsd) # 寻找色块 在ROIS中寻找色块，获取ROI中色块的中心区域与是否有色块的信息 def find_blobs_in_rois(self, img): canvas = img.copy() roi_blobs_result = {} # 在各个ROI中寻找色块的结果记录 for roi_direct in self.ROIS.keys(): roi_blobs_result[roi_direct] = { 'cx': 0, 'cy': 0, 'w': 0, 'blob_flag': False } for roi_direct, roi in self.ROIS.items(): blobs = canvas.find_blobs(self.LINE_COLOR_THRESHOLD, roi=roi, merge=True) if len(blobs) != 0: largest_blob = max(blobs, key=lambda b: b.pixels()) if largest_blob.area() > 1000: roi_blobs_result[roi_direct]['cx'] = largest_blob.cy() roi_blobs_result[roi_direct]['cy'] = largest_blob.cx() roi_blobs_result[roi_direct]['w'] = largest_blob.h() roi_blobs_result[roi_direct]['blob_flag'] = True x, y, width, height = largest_blob[:4] img.draw_rectangle((x, y, width, height), color=(255)) else: # blobs=canvas.find_blobs(LINE_COLOR_THRESHOLD, roi=roi, merge=True, pixels_area=10) continue blobs_baise = canvas.find_blobs(self.LINE_COLOR_BAISE, roi=(0, 0, 60, 240), merge=True) # 车载摄像头屏幕下部找白色#宽度200修改为240#用途寻卡，y示例中40修改为0 blobs_dixing = canvas.find_blobs(self.LINE_COLOR_BAISE, roi=(125, 0, 60, 240), merge=True) # 车载摄像头屏幕中间找白色 #宽度200修改为240，y示例中40修改为0 blobs_zuo = canvas.find_blobs(self.LINE_COLOR_THRESHOLD, roi=(0, 0, 180, 50), merge=True) # 车载摄像头屏幕左下部找黑色 blobs_you = canvas.find_blobs(self.LINE_COLOR_THRESHOLD, roi=(0, 190, 180, 50), merge=True) # 车载摄像头屏幕右下部找黑色 if len(blobs_baise) != 0: print("*********进入循环第1步*******") largest_baise = max(blobs_baise, key=lambda b: b.pixels()) wx, wy, wwidth, wwheight = largest_baise[:4] arc = wwidth * wwheight if arc >= 11000: print("*********进入循环第2步*******") if len(blobs_dixing) != 0: print("*********进入循环第3步*******") largest_dixing = max(blobs_dixing, key=lambda b: b.pixels()) wx, wy, wwidth, wwheight = largest_dixing[:4] arc = wwidth * wwheight if arc >= 11000: print('kapian') # 首先中部区域识别到白色进入判断地形还是卡片，接着判断下部，如果为白色判断为卡片。 self.UsartSend(self.data_format_wrapper(0, 1, 0)) # 地形停止命令 if len(blobs_zuo) != 0 and len(blobs_you) != 0: print("ka十字路口") self.UsartSend(self.data_format_wrapper(1, 1, 0)) # 地形停止命令 else: print('dixing') print(roi_blobs_result) # 返回的是黑色色块，各区域中心位置 self.UsartSend(self.data_format_wrapper(0, 1, 0)) # 地形停止命令 return roi_blobs_result # 返回的是黑色色块，各区域中心位置 # 计算偏转状态值 def state_deflection_angle(self, roi_blobs_result): ''' 说明：偏转状态值返回 ''' # ROI区域权重值 # ROIS_WEIGHT = [1, 1, 1, 1] ROIS_WEIGHT = [1, 0, 0, 1] state_crossing = False deflection_angle = 0 # 偏转角 down_center = 0 # 中下值 center_num = 0 # 中间值 # print(roi_blobs_result) ROIS_WEIGHT_num=0 # 偏转值计算，ROI中心区域X值 centroid_sum = roi_blobs_result['up']['cx'] * ROIS_WEIGHT[0] + roi_blobs_result['middle_up']['cx'] * ROIS_WEIGHT[1] \ + roi_blobs_result['middle_down']['cx'] * ROIS_WEIGHT[2] + roi_blobs_result['down']['cx'] * ROIS_WEIGHT[3] if roi_blobs_result['up']['blob_flag']: center_num += roi_blobs_result['up']['cx'] *ROIS_WEIGHT[0] ROIS_WEIGHT_num+=ROIS_WEIGHT[0] if roi_blobs_result['middle_up']['blob_flag']: center_num += roi_blobs_result['middle_up']['cx'] *ROIS_WEIGHT[1] ROIS_WEIGHT_num+=ROIS_WEIGHT[1] if roi_blobs_result['middle_down']['blob_flag']: center_num += roi_blobs_result['middle_down']['cx'] *ROIS_WEIGHT[2] ROIS_WEIGHT_num+=ROIS_WEIGHT[2] if roi_blobs_result['down']['blob_flag']: center_num += roi_blobs_result['down']['cx'] *ROIS_WEIGHT[3] ROIS_WEIGHT_num+=ROIS_WEIGHT[3] center_pos = center_num / ROIS_WEIGHT_num deflection_angle = (self.IMG_WIDTH / 2) - center_pos # 判断两侧ROI区域检测到黑色线 if roi_blobs_result['left']['blob_flag'] or roi_blobs_result['right']['blob_flag']: # 判断两侧ROI区域检测到黑色线处于图像下方1/3处 if roi_blobs_result['left']['cy'] <= ((self.IMG_HEIGHT / 3)) or roi_blobs_result['right']['cy'] <= ( (self.IMG_HEIGHT / 3)): # 当最下方ROI区域的黑线宽度大于140像素（检测到路口） if roi_blobs_result['down']['w'] > 140: down_center = 1 # 自行修改处 判断识别到十字路口 print("输出了十字路口标识") return down_center, state_crossing, deflection_angle # 控制舵机 def Tise_servo(self, angle): # 判断舵机控制方向 if angle < 0: # 限制舵机角度，防止过大损坏舵机 if angle > 80: angle = 80 angle = -angle elif angle > 0: # 限制舵机角度，防止过大损坏舵机 if angle > 35: angle = 35 angle = angle # PWM通过定时器配置，接到IO17引脚 tim_pwm = Timer(Timer.TIMER0, Timer.CHANNEL0, mode=Timer.MODE_PWM) S1 = PWM(tim_pwm, freq=50, duty=0, pin=17) S1.duty((angle + 90) / 180 * 10 + 2.5) def Servo(self, data): ''' 功能：180度舵机：angle:-90至90 表示相应的角度 360连续旋转度舵机：angle:-90至90 旋转方向和速度值。 【duty】占空比值：0-100 ''' angle = data[5] # 判断舵机控制方向 if data[4] == ord('-'): # 限制舵机角度，防止过大损坏舵机 if angle > 80: angle = 80 angle = -angle elif data[4] == ord('+'): # 限制舵机角度，防止过大损坏舵机 if angle > 35: angle = 35 angle = angle # PWM通过定时器配置，接到IO17引脚 tim_pwm = Timer(Timer.TIMER0, Timer.CHANNEL0, mode=Timer.MODE_PWM) S1 = PWM(tim_pwm, freq=50, duty=0, pin=17) S1.duty((angle + 90) / 180 * 10 + 2.5) # 检验校验和 def verify_checksum(self, data): if len(data) != 8: return False # 计算校验和：data[2]、data[3]、data[4] 和 data[5] 的和，然后对256取模 calculated_checksum = (data[2] + data[3] + data[4] + data[5]) % 256 # 比较计算出的校验和与data[6]是否相等 if calculated_checksum == data[6]: return True else: return False # 串口发送 def UsartSend(self, str_data): ''' 串口发送函数 ''' print(str_data) self.uart.write(str_data) # 判断符号 def get_symbol(self, num): ''' 根据数值正负，返回数值对应的符号 正数： ‘+’， 负数‘-’ 主要为了方便C语言解析待符号的数值。 ''' print("num = ", num) if num >= 0: return ord('+') else: return ord('-') # 封装数据 def data_format_wrapper(self, down_center, state_crossing, deflection_angle): ''' 根据通信协议封装循迹数据 TODO 重新编写通信协议 与配套C解析代码 ''' send_data = [ 0x55, 0x02, 0x91, down_center, # 底部色块中心是否在中点附近#底部色块十字路口 1 if state_crossing else 0, # 是否越障 无用 self.get_symbol(deflection_angle), # 偏航角符号 print输出是43 + 45- +向左转调整 -向右转调整 abs(int(deflection_angle)), # 偏航角 0xbb] return bytes(send_data) # 运行程序 myMain = Mainlen() myMain.startMain()