# Flock patent US 10,795,933 - object-based search and object fingerprinting > *Audited source extract. The dossier was used only as a lead; the underlying source was independently acquired and checked on 2026-07-20.* ## Source metadata - **Publisher:** United States Patent and Trademark Office - **Original dossier URL:** https://patents.justia.com/patent/10795933 - **Resolved/canonical URL used:** https://image-ppubs.uspto.gov/dirsearch-public/print/downloadPdf/10795933 - **Publication date:** 2020-10-06 - **Underlying event or version:** not separately stated - **Archived:** 2026-07-20, from unauthenticated public access; no cookies, tokens, or login state retained - **Wayback snapshot:** save failed on 2026-07-20 after the archive service returned HTTP 403; retry pending ## Verification verdict **supported with narrower wording** — US 10,795,933 B1 supports the disclosed object-search and fingerprinting architecture. Product pages or deployment records are still required to establish current implementation and scope. ## Claim boundary This source is authoritative only within the source class and limitations stated above. ## Extract **Extracted-text lines 1-3:** > US10795933B1 - System and method for object based query of video content captured by a dynamic surveillance network > - Google Patents > US10795933B1 - System and method for object based query of video content captured by a dynamic surveillance network **Extracted-text lines 335-339:** > G06V20/46 > — > Extracting features or characteristics from the video content, e.g. video fingerprints, representative shots or key frames > G > — **Extracted-text lines 491-495:** > is “readable” if a symbology associated with an identified object may be accurately decoded or interpreted. > two identifiable and readable objects captured in separate video frames > may be considered equivalent, and thus the same object, if an object similarity probability calculation exceeds a given threshold. > a probability calculation > may consider the statistical significance of similarity among the objects' respective fingerprints, relative locations, object neighborhood, etc. **Extracted-text lines 650-654:** > the object detection module 154 > may leverage neural networks 507 , 509 (See FIGS. 5A-5B ) in order to classify identified objects and determine the nature of aspects of the object (e.g., color, type, size, etc.). The object detection module 154 may then store the analyzed video in the DSOQ database 120 in such a way that portions of the video footage are associated with, and thus can be queried by, the objects identified therein. > the search module 156 > may take user inputs from I/O and Display 152 and query DSOQ database 120 in order to identify relevant footage for return to Display 152 . > the inputs from I/O and Display 152 **Wayback failure**: The bounded Save Page Now pass stopped after the archive service returned HTTP 403 on the first request. This source was not individually retried; retry pending. ## Notes - Tier: 2 — Tier 2 primary public record. - The excerpt is bounded to the claims audited here; consult the preserved original for full context. - Cited by: see `citing_pages:` frontmatter; populated after wiki integration.