Video surveillance system, video processing apparatus, video processing method, and video processing program
A video processing apparatus includes a video analyzer that analyzes video data captured by a surveillance camera, detects an event belonging to a specific category, and outputs a detection result, a display controller that displays, together with a video of the video data, a category setting screen for setting a category of an event included in the video, and a learning data accumulator that accumulates, as learning data together with the video data, category information set in accordance with an operation by an operator to the category setting screen. The video analyzer performs learning processing by using the learning data accumulated in the learning data accumulator.
1. A video processing apparatus comprising:
a display controller implemented at least by hardware and that displays, together with a video of the video data captured by a surveillance camera, a category setting screen for setting a category included in the video, the category indicating a dangerous goods to be detected or operation to be detected,
a learning data accumulator implemented at least by a processor executing code from a memory and that accumulates, as learning data together with the video data, information of the category set in accordance with an operation by an operator to the category setting screen; and
a first video analyzer implemented at least by the processor executing the code from the memory that performs learning processing by using the learning data accumulated in said learning data accumulator, detects a specific category by analyzing said video data, and outputs a detection result,
wherein said display controller causes the operator to input information necessary for newly generating a new category associated with newly captured video data in the category setting screen, when receiving form the operator that the new category is included in the newly captured video data and cannot be specified by the operator,
when the new category is newly generated, a second video analyzer, at least having an algorithm different from that of the first video analyzer, for detecting the new category is newly generated.
2. The video processing apparatus according to claim 1 , wherein said display controller displays a plurality of categories associated with the video data in the category setting screen so as to be able to modify or add the plurality of categories.
3. The video processing apparatus according to claim 1 , wherein said display controller displays a plurality of categories associated with the video data in the category setting screen so as to be able to select the plurality of categories.
4. The video processing apparatus according to claim 1 , wherein said learning data accumulator accumulates, as the information of the category, a category name, a video capturing time, a region including the dangerous goods or problematic behaviors, and operator information in said learning data accumulator.
5. A video surveillance system comprising:
a video data accumulator implemented at least by a processor that executes code from a memory and that accumulates video data captured by a surveillance camera;
a display controller implemented at least by hardware that displays, together with a video of the video data accumulated in said video data accumulator, a category setting screen for setting a category included in the video, the category indicating a dangerous goods to be detected or operation to be detected;
a first video analyzer implemented at least by the processor that executes the code from the memory and that performs learning processing by using the learning data accumulated in said learning data accumulator, detects a specific category by analyzing said video data, and outputs a detection result; and
a learning data accumulator implemented at least by a processor executing code from a memory and that accumulates, as learning data together with the video data, information of the category set in accordance with an operation by an operator to the category setting screen,
wherein said first video analyzer performs learning processing by using the learning data accumulated in said learning data accumulator,
said display controller causes the operator to input information necessary for newly generating a category associated with newly captured video data in the category setting screen, when receiving form the operator that the new category is included in the newly captured video data and cannot be specified by the operator,
when the new category is newly generated, a second video analyzer, at least having an algorithm different from that of the first video analyzer, for detecting the dangerous goods or operation belonging to the new category is newly generated.
6. A video processing method comprising:
displaying, together with a video of the video data, a category setting screen for setting a category included in the video, the category indicating the dangerous goods to be detected or operation to be detected;
accumulating, as learning data together with the video data, information of the category set in accordance with an operation by an operator to the category setting screen;
performing learning processing of the video analysis module by using the learning data accumulated in the accumulating;
analyzing video data captured by a surveillance camera by using a video analysis module, detecting a specific category, and outputting a detection result; and
causing the operator to input information necessary for newly generating a new category associated with newly captured video data in the category setting screen, when receiving form the operator that the new category is included in the newly captured video data and cannot be specified by the operator;
wherein when the new category is newly generated, a second video analyzer, at least having an algorithm different from that of the first video analyzer, for detecting the new category is newly generated.
7. A non-transitory computer readable medium storing a video processing program for causing a computer to execute a method, comprising:
displaying, together with a video of the video data, a category setting screen for setting a category included in the video, the category indicating the dangerous goods to be detected or operation to be detected;
accumulating, as learning data together with the video data, information of the category set in accordance with an operation by an operator to the category setting screen;
performing learning processing of the video analysis module by using the learning data accumulated in the accumulating;
analyzing video data captured by a surveillance camera by using a video analysis module, detecting a specific category, and outputting a detection result; and
causing the operator to input information necessary for newly generating a new category associated with newly captured video data in the category setting screen, when the new category is included in the newly captured video data and cannot be specified by the operator;
wherein when the new category is newly generated, a second video analyzer, at least having an algorithm different from that of the first video analyzer, for detecting the new category is newly generated.