交通标志识别Python+TensorFlow实现（QT界面+WEB界面）

介绍

交通标志识别系统，采用Python+TensorFlow构建神经网络，通过对数据集图像的训练，得到模型，然后采用QT构建桌面端可视化操作软件，Django构建网页端WEB可视化操作平台。可识别50多种常见的交通标志。

QT版展示

网页版展示

代码

def upload_img(request):
    # 图片上传
    file = request.FILES.get('file')
    file_name = file.name
    with open(os.path.join(settings.MEDIA_ROOT, file_name), 'wb') as f:
        for chunk in file.chunks():
            f.write(chunk)
    upload_url = request.build_absolute_uri(settings.MEDIA_URL + file_name)
    try:
        ImageCheck.objects.create(file_name=file_name, file_url=upload_url)
    except ImageCheck:
        return restful.server_error(message='数据库发生错误！')
    return restful.ok(data={'url': upload_url})


def check_img(request):
    # 图片检测
    image_url = request.POST.get('img_url')
    if not image_url:
        return restful.params_error(message='缺少必要的参数image_url')
    image_name = image_url.rsplit('/')[-1]
    image_path = os.path.join(settings.MEDIA_ROOT, image_name)
    pred_name = check_handle(image_path)

    try:
        obj = ImageCheck.objects.filter(file_name=image_name).last()
        obj.check_result = pred_name
        obj.save()
    except:
        return restful.server_error(message='数据库发生错误')
    return restful.ok(data={'flower': pred_name, 'chance': str(pred_name) or '0'})

需要联系q：2784127853