Priority Based Image Processing Methods
An image recognition approach employs both computer generated and manual image reviews to generate image tags characterizing an image. The computer generated and manual image reviews can be performed sequentially or in parallel. The generated image tags may be provided to a requester in real-time, be used to select an advertisement, and/or be used as the basis of an internet search. In some embodiments generated image tags are used as a basis for an upgraded image review. A confidence of a computer generated image review may be used to determine whether or not to perform a manual image review.
1 . A method of processing an image, the method comprising:
receiving an image from an image source;
distributing the image to an automated image identification system;
receiving a computer generated review from the automated image identification system, the computer generated review including one or more image tags assigned to the image by the automated image identification system and a measure of confidence, the measure of confidence being a measure of confidence that the image tags assigned to the image correctly characterize contents of the image;
assigning a priority to the image based, at least in part, on the measure of confidence;
determining that the image should be manually tagged based on the priority;
posting the image for manual review to a destination, the destination including a display device of a human image reviewer; and
receiving a manual image review of the image from the destination, the image review including one or more image tags assigned to the image by the human image reviewer, the one or more image tags assigned by the human image reviewer characterizing contents of the image.
2 . The method of claim 1 , further comprising extracting the image from a webpage.
3 . The method of claim 1 , further comprising selecting an advertisement based on the one or more tags.
4 . The method of claim 3 , wherein the selection of the advertisement is based on a real-time auction of the one or more image tags.
5 . The method of claim 1 , wherein assigning the priority is further based on a number of times the image is viewed.
6 . The method of claim 1 , wherein assigning the priority is further based on a sum of a number of times the image is viewed on multiple webpages.
7 . The method of claim 1 , wherein the automated image identification system is configured to determine if a copy of the image received from a different image source has already been tagged.
8 . A method of processing an image, the method comprising:
receiving an image from an image source;
automatically determining a priority to the image using a microprocessor;
determining how the image should be tagged based on the priority;
tagging the image to produce one or more tags, the one or more tags characterizing contents of the image; and
storing the image and the one or more tags in a data structure.
9 . The method of claim 8 , wherein the priority is based on information regarding the source of the image.
10 . The method of claim 9 , wherein the information regarding the source includes how often one or more webpages including the image are viewed.
11 . The method of claim 9 , wherein the information regarding the source includes a type of webpage including the image.
12 . The method of claim 9 , wherein the information regarding the source includes how often advertisements assigned to the image are clicked.
13 . The method of claim 9 , wherein the information regarding the source includes text included on a same webpage as the image.
14 . The method of claim 9 , wherein the information regarding the source includes how often the image is viewed.
15 . The method of claim 9 , wherein the information regarding the source includes a ranking of a webpage including the image.
16 . The method of claim 8 , wherein the priority is based on a number of webpages that include the image.
17 . The method of claim 8 , wherein the priority is based on a position of the image within a webpage.
18 . The method of claim 8 , further comprising tagging the image using an automatic identification system, to produce one or more additional tags; and wherein the priority is based on a confidence that the additional tags produced using the automatic identification system accurately characterize contents of the image.
19 . The method of claim 8 , further comprising tagging the image using an automatic identification system, to produce one or more additional tags characterizing the image; and wherein the priority is further based an estimated value of the one or more additional tags produced using the automatic identification system.
20 . The method of claim 8 , wherein tagging the image includes sending the image to a human reviewer via a computer network.
21 . The method of claim 8 , wherein determining how the image should be tagged includes selecting between manual or automated tagging of the image.
22 . The method of claim 21 , wherein images that are frequently viewed are selected for manual tagging and images that are viewed relatively less frequently are selected for automated tagging.
23 . The method of claim 8 , wherein tagging the image includes both tagging the image using an automatic identification system and sending the image to a human reviewer for manual tagging.
24 . The method of claim 8 , wherein images must have a priority above a predetermined threshold to be manually tagged by a human image reviewer.