{
  "$type": "site.standard.document",
  "coverImage": {
    "$type": "blob",
    "ref": {
      "$link": "bafkreibgc3mtjnvrhio2pzjua2kl52dlg5thrzv6hxlqzdn46dxosricxq"
    },
    "mimeType": "image/png",
    "size": 120573
  },
  "description": "A violation inspection system based on visual sensing of a self-driving vehicle and a method thereof are disclosed. The violation inspection system receives traffic images captured by the self-driving vehicle, identifies image objects by artificial intelligence, continuously detects a movement…",
  "path": "/patents/1343636",
  "publishedAt": "2023-05-18T00:00:00.000Z",
  "site": "at://did:plc:oql6ds5vnff4ugar6rruliwd/site.standard.publication/3mn3ohu7oxx5w",
  "tags": [
    "B60W60/001",
    "SQ Technology (Shanghai) Corporation"
  ],
  "textContent": "A violation inspection system based on visual sensing of a self-driving vehicle and a method thereof are disclosed. The violation inspection system receives traffic images captured by the self-driving vehicle, identifies image objects by artificial intelligence, continuously detects a movement status of each of the image objects, to generate image object dynamics; when the image object matches one of a trigger object factor and an association object factor, the image object and the image object dynamics thereof are inputted to an AI-based situation recognition model, to output a current situation, and when the current situation matches a violation situation template, the image objects corresponding to the trigger object factor and the association object factor corresponding to the violation situation template, and a violation item are outputted, so as to achieve the technical effect of improving accuracy and diversity in violation behavior determination.",
  "title": "Violation Inspection System Based on Visual Sensing of Self-Driving Vehicle and Method Thereof"
}