Get Lite Deeplearning Object
Function:
go.get_arm_camera_obj()Description: get data of the object detected by Magician Lite
- Required parameter: null
- Return: {count, dl_obj[count]}
- count: number of objects
- dl_obj [i]={x, y, w, h, ID}: parameter of object in visual coordinate system
- x: x-axis coordinate of the center point of image recognition box
- y: y-axis coordinate of the center point of image recognition box
- w: width of image recognition box
- h: height of image recognition box
- ID: categories of object
When the detection mode of arm camera is block detection, the categories that the ID corresponds to are as follows:
| ID | Category (in Chinese) | Category (in English) |
|---|---|---|
| 0 | 红 | red |
| 1 | 黄 | yellow |
| 2 | 蓝 | blue |
| 3 | 绿 | green |
When the detection mode of arm camera is express and fresh detection, the categories that the ID corresponds to are as follows:
| ID | Category (in Chinese) | Category (in English) |
|---|---|---|
| 0 | 橙子 | Orange |
| 1 | 白菜 | Cabbage |
| 2 | 矿泉水 | Water Bottle |
| 3 | 国内快递 | Domestic Express |
| 4 | 国际快递 | International Express |
| 5 | 同城快递 | City Express |
Example
datalist = go.get_arm_camera_obj() # Get information detected by Lite if datalist["count"]: # If there are blocks on box grab_ind = randint(0,int(datalist["count"]-1)) # select a block grab_obj = datalist["dl_obj"][grab_ind] # Get block information x_c = grab_obj["x"]+grab_obj["w"]/2 # Calculate the coordinate of block center y_c = grab_obj["y"]+grab_obj["h"]/2Get data of the object detected by Magician Lite.