System and method for estimating/determining the date of a photo
Undated photos are organized by estimating the date of each photo. The date is estimated by building a model based on a set of reference photos having established dates, and comparing image characteristics of the undated photo to the image characteristics of the reference photos. The photo characteristics can include hues, saturation, intensity, contrast, sharpness and graininess as represented by image pixel data. Once the date of a photo is estimated, it can be tagged with identifying information, such as by using the estimated date to associate the photo with a node in a family tree.
1. A method for estimating an image capture date of a photo, comprising:
identifying, by a processor, image characteristics of an undated photo;
comparing, by the processor, the image characteristics of the undated photo to image characteristics relating to a plurality of reference photos, each of the plurality of reference photos having an established date, wherein the compared image characteristics comprise graininess and wherein graininess is represented by frequency domain data;
estimating, by the processor, an image capture date of the undated photo based on a similarity of the image characteristics of the undated photo to one or more of the image characteristics of the reference photos;
displaying at a user interface, by the processor, the undated photo;
displaying at the user interface, by the processor, a data structure having a plurality of data nodes, each data node representing one or more data records having a subject associated with one or more dates;
identifying at the displayed data structure, by the processor, one or more data nodes that have a relationship to the undated photo based on the estimated image capture date of the undated photo and the dates associated with the subject of the one or more data nodes;
displaying, by the processor, a tagging data entry window at the user interface for receiving tagging information from a user viewing the undated photo and the displayed data structure; and
receiving, by the processor, tagging information at the tagging data entry window at the user interface concerning the undated photo based on the identified one or more data nodes.
2. The method of claim 1 , wherein the image characteristics are characteristics of pixel data resulting from digital processing of an image in the undated photo.
3. The method of claim 2 , wherein the image characteristics further comprise sharpness, and wherein sharpness is represented by frequency domain data.
4. The method of claim 2 , wherein the image characteristics further comprise color, and wherein color is represented by an HSI (Hue, Saturation, Intensity) color model data.
5. The method of claim 2 , wherein the pixel data comprises:
pixel data transformed into a Fourier series representing frequency domain data.
6. The method of claim 1 , further comprising creating an estimation model for estimating image capture dates by:
determining, by the processor, the image characteristics of the reference photos;
assigning, by the processor, the image characteristics of the reference photos to reference dates; and
developing, by the processor, rules for the estimation model for associating image characteristics of undated photos to image characteristics of the reference photos, based on the image characteristics of the reference photos assigned to reference dates;
wherein the step of comparing the image characteristics of the undated photo comprises applying, by the processor, the image characteristics of the undated photo to the rules developed for the estimation model.
7. The method of claim 1 , wherein graininess is determined by the frequency of Fourier series signals represented across the image of the undated photo.
8. The method of claim 1 , wherein the data structure is a family tree, and wherein the subject of each node is a person.
9. The method of claim 8 , wherein the undated photo and the family tree are displayed together at the user interface.
10. The method of claim 1 , wherein the data structure having a plurality of data nodes comprises a family tree, with each data node in the family tree associated with a person, and wherein the method further comprises:
storing the undated photo as a data record at the one or more identified data nodes that have a relationship to the undated photo.
11. The method of claim 1 , wherein identifying image characteristics of the undated photo comprises calculating an average graininess of the undated photo based on the frequency of Fourier series signals represented across the image in the undated photo.
12. The method of claim 1 , wherein the plurality of reference photos comprise photos stored in a database system maintaining family trees and having information associated with dates that the stored photos were taken.