You can not select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.

5.8 KiB

This file contains ambiguous Unicode characters!

This file contains ambiguous Unicode characters that may be confused with others in your current locale. If your use case is intentional and legitimate, you can safely ignore this warning. Use the Escape button to highlight these characters.

一、环境信息

水口口岸二桥货场AR全景可视化监管平台 admin 123456

二、问题记录

  1. 7+4 入境4缺少车底相机192.168.33.99 Ip不存在
  2. 7+4 出境7缺少车道头相机未激活
  3. 车辆通关计算不准确
  4. 不要用AR提供的设备管理功能设备操作必须使用权限系统
  5. 相机有变动IP和位置需要重新配置起点、终点、沿途类型
  6. 数据一致性问题:有验放没有车牌识别,有车牌识别没有验放

三、FAQ

3.1 如何添加关联相机

权限系统中增加关联回放设备idacc_t_sb表主键 !Pasted image 20250627152148.png

四、常用SQL

  1. 查询车牌识别、验放推送记录
select (case  
            when type = 1 then '车牌识别'  
            when type = '2' then '车辆验放'  
            when type = '3' then '登临检查' end) as type,  
       create_time,  
       device_name,  
       plate,  
       data  
from ar_t_vehicle_data_receive_log  
where plate = '12H02835'  
order by create_time;
# 过滤车牌识别
select device_name, plate, create_time, data  
from ar_t_vehicle_data_receive_log  
where plate = '桂P13077'  
  and type = 1  
  and device_id is not null  
  and create_time>='2025-06-24'  
order by create_time;
# 轨迹表
with io_record as (select * from ar_t_sleuth_io where vehicle_plate = '桂P13077'),  
     device as (select * from acc_t_sb)  
select a.id,a.vehicle_io_id, a.vehicle_plate, a.snap_time, b.SBM, a.device_id  
from io_record a  
         left join device b on a.device_id = b.id where a.snap_time>='2025-06-24'  
order by a.snap_time,a.vehicle_io_id;
# 卡口验放
with release_record as (select * from ar_t_vehicle_io_release_detail where vehicle_plate = '桂KQ6619'),  
     device as (select * from acc_t_sb)  
select a.sleuth_io_id, b.SBM, a.device_id, a.vehicle_plate, a.vehicle_pass_time  
from release_record a  
         left join device b on a.device_id = b.ID  
where a.vehicle_pass_time >= '2025-06-28'  
order by a.vehicle_pass_time;

五、任务

一、同步验放数据 (完成)

  1. 使用嵌入式debezium-engine同步验放表、通道表
  2. 开发视图用于查询

二、修正车牌识别数据(有车牌识别没有验放数据、有验放数据没有车牌识别)

  1. (有验放数据没有车牌识别)在接收到验放数据,校验是否有车牌识别,如果没有则通过验放数据构造一条车牌识别数据
  2. (有车牌识别没有验放数据)无需处理

三、车辆通关状态查询重构

  1. 根据步骤一实现的视图重构 车辆通关状态查询 模块逻辑(完成)
  2. 修改现有查询接口(完成)

四、部署 hertzbeat完成 http://192.168.20.17:1157/ admin/Max@123456

五、实现GIS业务表、权限系统业务表定时备份 (完成) 192.168.20.17 /opt/db_backup.sh 输出到 /home/maxvison/backup

六、优化项

  1. 3D地图打开相机视频改为弹窗防止重复地图加载 前端反馈不好实现已改为新开tab页

七、设计

7.1 合并验放数据

  1. 通过验放数据构建视图
create view vw_release_t_channel_records as  
select a.domestic_no as vehivle_plate,-- 车牌号  
       a.io_type,-- 进出口标识 I 进口 O 出口  
       a.channel_name,-- 通道名称  
       a.channel_tag,-- 通道标识(s,开始e,结束p,中间)  
       d.SBM                                        as acc_camera_name,-- AR系统中的相机名称  
       DATE(a.in_time)                              as in_date,-- 入通道日期  
       DATE(a.out_time)                             as out_date,-- 出通道日期  
       a.in_time, -- 入通道时间  
       a.out_time,-- 出通道时间  
       TIMESTAMPDIFF(MINUTE, a.in_time, a.out_time) as time_diff_minute,-- 进出通道时间差(分钟)  
       b.customs_channel_no,-- 海关标识通道号  
       b.customs_area_id,-- 海关标识区域ID  
       a.bj_state,-- 边检放行状态 NN不开启YP放行CP查验EP退车TP应急放行  
       a.hg_state,-- 海关放行状态 NN不开启YP放行CP查验EP退车TP应急放行  
       a.weighbridge,-- 地磅KG  
       a.final_state,-- 最终状态 0抬杆放行,1退车,2人工抬杆  
       a.channel_no,-- 通道编号  
       b.guid                                       as channel_id,-- 通道ID  
       a.channel_type,-- 通道类型  
       b.business_type                              as channel_business_type,-- 通道业务类型 1一般贸易2 边民互市3核辐射测温  
       c.camera_id,-- 相机ID  
       c.camera_name,-- 验放系统中的相机名称  
       c.camera_url -- RTSP 地址  
from release_t_channel_records a  
         left join release_t_channel b on a.channel_no = b.channel_no  
         left join (select * from release_t_camera where camera_type = '5') c on b.guid = c.channel_id  
         left join acc_t_sb d on c.camera_id = d.ID;
  1. 历史数据处理 按照车牌分组26号以前数据有问题
with release_record as (select *  
                        from vw_release_t_channel_records  
                        where in_date >= '2025-06-26'  
                        order by in_date, in_time)  
select *  
from release_record  
where vehivle_plate = '桂FB5077';

按照IO类型分组分组之后再校验channel_tag将数据合并如下图所示可以将数据合并为3条 !Pasted image 20250701085900.png 3. 新数据处理 方法一定时任务查询近3天未结束流程的数据更新流程。有可能3天都断网拿不到最新的数据。等回复后又过了3天的周期。或者每晚加一个定时任务进行全量构建 方法二监听CDC任务的insert然后更新通关流程