公开(公告)号：GB2565211A

公开(公告)日：2019-02-06

申请号：GB201809525

申请日：2018-06-11

Applicant: IBM

Inventor： EVELYN ANDERSON ,  NATALIE BROOKS POWELL ,  KRISTEN CONLEY ,  MARTIN KEEN

IPC: G06K9/00 ,  A01G7/06 ,  G06Q50/02

Abstract: A computer implemented method for providing a clinical service for plants comprises obtaining an input including an image of a plant 101 from a camera 105 configured by a user, and identifying the plant species by image recognition 119. The image is analyzed for symptoms 130, and candidate diagnoses 140, each with a respective confidence score, are selected from a diagnostic repository 113 according to the symptoms. The candidate diagnoses are each mapped to a respective treatment regimen 150 and delivered 199 according to a preconfigured treatment mode 125, which may be manual or automated. The input may include a geographical location, a timestamp, other plant images, environmental 185, weather 115 and soil data 117. When the treatment mode is automated, the treatment regimen corresponding to the diagnosis with the highest confidence score may be performed using preconfigured measures selected from ordering a treating substance or equipment, controlling a water supply, controlling lighting and/or controlling temperature. A system 100 includes one or more processors in communication with a memory, and executes program instructions to perform the method.

公开(公告)号：GB2600281A

公开(公告)日：2022-04-27

申请号：GB202200555

申请日：2020-05-21

Applicant: IBM

Inventor： MICHAEL BENDER ,  MICHAEL SHUTE ,  SIDDHARTHA SOOD ,  EVELYN ANDERSON

IPC: G06K9/00

Abstract: A method, computer program product, and a system where a processor(s) obtains a video from a user, via a client, and segments the video into temporal shots that comprise a timeline of the video. The processor(s) cognitively analyze the video, by applying an image recognition algorithm to identify image entities in each temporal shot of the video and by applying a data structure comprising a user profile of the user to the temporal shots, to identity personal entities in each temporal shot of the video. The program code generates a search index for the video, utilizing the user entities (image entities and personal entities), where each entry of the search index is a given user entity and a linkage to a given temporal shot and the linkage indicates a location of the given user entity in the timeline of the video.