Method and apparatus for automatic visual detection, recording, and retrieval of events

Abstract

Disclosed are methods and apparatus for automatic visual detection of events, for recording images of those events and retrieving them for display and human or automated analysis, and for sending synchronized signals to external equipment when events are detected. An event corresponds to a specific condition, among some time-varying conditions within the field of view of an imaging device, that can be detected by visual means based on capturing and analyzing digital images of a two-dimensional field of view in which the event may occur. Events may correspond to rare, short duration mechanical failures for which obtaining images for analysis is desirable. Events are detected by considering evidence obtained from an analysis of multiple images of the field of view, during which time moving mechanical components can be seen from multiple viewing perspectives.

1 . A method for automatic visual detection, recording, and retrieval of an event, comprising: 
 capturing a sequence of frames, each frame in the sequence of frames comprising an image of a field of view in which the event occurs, the event corresponding to a specific condition, among time-varying conditions within the field of view, that can be detected by visual means;    analyzing each frame of the sequence of frames to obtain evidence that the event has occurred and to identify at least one event frame, the analysis of the at least one event frame revealing sufficient evidence that the event has occurred;    choosing a plurality of selected frames from the sequence of frames, such that the step of choosing to include a frame from the sequence of frames in the plurality of selected frames is responsive to the position of the frame in the sequence of frames relative to the at least one event frame;    recording the plurality of selected frames in a memory;    receiving commands to retrieve frames from the recorded plurality of selected frames; and    retrieving frames from the recorded plurality of selected frames in response to the commands.    
   
   
       2 . The method of  claim 1 , wherein: 
 the commands are responsive to an activity of a human user; and    the retrieved frames are displayed for the human user.    
   
   
       3 . The method of  claim 1 , wherein at least one of the frames in the plurality of selected frames occurs prior to the at least one event frame in the sequence of frames.  
   
   
       4 . The method of  claim 1 , wherein at least one of the frames in the plurality of selected frames occurs after the at least one event frame in the sequence of frames.  
   
   
       5 . The method of  claim 1 , wherein: 
 at least one of the frames in the plurality of selected frames occurs prior to the at least one event frame in the sequence of frames; and    at least one of the frames in the plurality of selected frames occurs after the at least one event frame in the sequence of frames.    
   
   
       6 . The method of  claim 1 , wherein: 
 the commands are responsive to an automated image analysis system; and    the retrieved frames are analyzed by the automated image analysis system.    
   
   
       7 . The method of  claim 2 , further comprising: 
 displaying a plurality of retrieved frames as a corresponding plurality of low-resolution “thumbnail” images; and    displaying at high-resolution at least one retrieved frame corresponding to at least one of the plurality of low-resolution “thumbnail” images.    
   
   
       8 . The method of  claim 1 , wherein the analyzing step further comprises: 
 analyzing each frame in the sequence of frames to determine a corresponding sequence of event detection weights, each event detection weight in the sequence of event detection weights comprising evidence that the event has occurred in the corresponding frame; and    analyzing the sequence of event detection weights to identify the at least one event frame, such that a combined evidence responsive to the event detection weights corresponding to the at least one event frame reveals sufficient evidence that the event has occurred.    
   
   
       9 . The method of  claim 8  wherein: 
 the at least one event frame comprises a plurality of event frames; and    the combined evidence is responsive to all of the event detection weights corresponding to the plurality of event frames.    
   
   
       10 . A system for automatic visual detection, recording, and retrieval of events, comprising: 
 a capture process that captures a sequence of frames, each frame in the sequence of frames comprising an image of a field of view in which the event occurs, the event corresponding to a specific condition, among time-varying conditions within the field of view, that can be detected by visual means;    an analysis process that analyzes each frame of the sequence of frames to obtain evidence that the event has occurred and to identify at least one event frame, the analysis of the at least one event frame revealing sufficient evidence that the event has occurred;    a selection process that chooses a plurality of selected frames from the sequence of frames, such that the selection process chooses to include a frame from the sequence of frames in the plurality of selected frames responsive to the position of the frame in the sequence of frames relative to the at least one event frame;    a memory that records the plurality of selected frames; and    an interface that receives commands to retrieve frames from the recorded plurality of selected frames and that retrieves frames from the recorded plurality of selected frames in response to the commands.    
   
   
       11 . The system of  claim 10 , wherein the interface includes a human-machine interface, and wherein: 
 the commands are responsive to an activity of a human user with the human-machine interface; and    the interface is adapted to display retrieved frames for the human user.    
   
   
       12 . The system of  claim 10 , wherein at least one of the frames in the plurality of selected frames occurs prior to the active frame in the sequence of frames.  
   
   
       13 . The system of  claim 10 , wherein at least one of the frames in the plurality of selected frames occurs after the active frame in the sequence of frames.  
   
   
       14 . The system of  claim 10 , wherein: 
 at least one of the frames in the plurality of selected frames occurs prior to the active frame in the sequence of frames; and    at least one of the frames in the plurality of selected frames occurs after the active frame in the sequence of frames.    
   
   
       15 . The system of  claim 10 , wherein the interface includes an automated image analysis system, and wherein: 
 the commands are responsive to the automated image analysis system; and    the retrieved frames are analyzed by the automated image analysis system.    
   
   
       16 . The system of  claim 11 , wherein the interface includes: 
 a display adapted to display a plurality of retrieved frames as a corresponding plurality of low-resolution “thumbnail” images, the display being adapted to allow selection, responsive to the human user, of at least one “thumbnail” image from the plurality of “thumbnail” images; and    a high-resolution display adapted to display, at high-resolution, at least one retrieved frame corresponding to the at least one “thumbnail” image.