K210M1 dock车牌识别

车牌识别

---

前言

先来看一下效果视频

K210 dock车牌识别

一、车牌识别

选用的是K210 M1Dock，实现用的是01的固件，效果确实可以，原本他们是用在canmv上的，我把固件烧录到了K210上，然后把模型放到K210里，同样可以运行。

二、模型文件和代码

1.模型文件

然后看一下固件

这个是中文字库

2.代码

import sensor, image, time, lcd
from maix import KPU, utils
import gc

lcd.init()
sensor.reset()                      # Reset and initialize the sensor. It will
                                    # run automatically, call sensor.run(0) to stop
sensor.set_pixformat(sensor.RGB565) # Set pixel format to RGB565 (or GRAYSCALE)
sensor.set_framesize(sensor.QVGA)   # Set frame size to QVGA (320x240)
sensor.skip_frames(time = 1000)     # Wait for settings take effect.
#sensor.set_hmirror(1)
#sensor.set_vflip(1)
clock = time.clock()                # Create a clock object to track the FPS.
#image.font_load(image.UTF8, 16, 16, '/sd/font/0xA00000_font_uincode_16_16_tblr.Dzk') # load chinese font file
image.font_load(image.UTF8, 16, 16, 0xA00000)

province_cn = ("皖沪津渝冀晋蒙辽吉黑苏浙京闽赣鲁豫鄂湘粤桂琼川贵云藏陕甘青宁新")
ads = ('A', 'B', 'C', 'D', 'E', 'F', 'G', 'H', 'J', 'K', 'L', 'M', 'N', 'P', 'Q', 'R', 'S', 'T', 'U', 'V', 'W', 'X', 'Y', 'Z', '0', '1', '2', '3', '4', '5', '6', '7', '8', '9')

anchor = (8.30891522166988, 2.75630994889035, 5.18609903718768, 1.7863757404970702, 6.91480529053198, 3.825771881004435, 10.218567655549439, 3.69476690620971, 6.4088204258368195, 2.38813526350986)
kpu = KPU()
kpu.load_kmodel("/sd/KPU/licenseplate_recognization/lp_detect.kmodel")
kpu.init_yolo2(anchor, anchor_num=5, img_w=320, img_h=240, net_w=320 , net_h=240 ,layer_w=20 ,layer_h=15, threshold=0.7, nms_value=0.3, classes=0)

lp_recog_kpu = KPU()
lp_recog_kpu.load_kmodel("/sd/KPU/licenseplate_recognization/lp_recog.kmodel")
#lp_recog_kpu.lp_recog_load_weight_data(0x600000, 1498500) # load after-process data
lp_recog_kpu.lp_recog_load_weight_data("/sd/KPU/licenseplate_recognization/lp_weight.bin") # load after-process data

RATIO = 0.16
def extend_box(x, y, w, h, scale):
    x1_t = x - scale*w
    x2_t = x + w + scale*w
    y1_t = y - scale*h
    y2_t = y + h + scale*h
    x1 = int(x1_t) if x1_t>1 else 1
    x2 = int(x2_t) if x2_t<320 else 319
    y1 = int(y1_t) if y1_t>1 else 1
    y2 = int(y2_t) if y2_t<256 else 255
    cut_img_w = x2-x1+1
    cut_img_h = y2-y1+1
    return x1, y1, cut_img_w, cut_img_h

lp_index_list = []

while 1:
    clock.tick()                    # Update the FPS clock.
    img = sensor.snapshot()
    kpu.run_with_output(img)
    dect = kpu.regionlayer_yolo2()
    fps = clock.fps()
    if len(dect) > 0:
        #print("dect:",dect)
        for l in dect :
            x1, y1, cut_img_w, cut_img_h= extend_box(l[0], l[1], l[2], l[3], scale=RATIO)
            lp_cut = img.cut(x1, y1, cut_img_w, cut_img_h)
            a=img.draw_rectangle(l[0],l[1],l[2],l[3], color=(255, 0, 0))
            lp_resize = lp_cut.resize(208,64)
            a=lp_resize.replace(vflip=0, hmirror=1)
            lp_resize.pix_to_ai()
            lp_recog_kpu.run_with_output(lp_resize)
            out = lp_recog_kpu.lp_recog()
            #print("out:",len(out))
            lp_index_list.clear()
            for n in out:
                max_score = max(n)
                index = n.index(max_score)
                lp_index_list.append(index)
            del (lp_cut)
            del (lp_resize)
            show_lp_str = "%s   %s-%s%s%s%s%s" %(province_cn[lp_index_list[0]], ads[lp_index_list[1]], ads[lp_index_list[2]],
                ads[lp_index_list[3]], ads[lp_index_list[4]], ads[lp_index_list[5]], ads[lp_index_list[6]])
            print(show_lp_str)
            a=img.draw_string(l[0], l[1]-18, show_lp_str, color=(255, 128, 0), scale=1)
    #img.replace(vflip=0, hmirror=1)
    a=img.draw_string(10, 0, "%2.1ffps" %(fps), color=(255, 255, 0), scale=1)
    lcd.display(img)
    # print("mem free:",gc.mem_free())
    # print("heap free:",utils.heap_free())
    gc.collect()

image.font_free()
kpu.deinit()
lp_recog_kpu.deinit()

完整的代码已经上传到资源里，可以自行下载。