Anomaly detection apparatus, method, system and program

1. An abnormality detection device comprising:

processing circuitry configured to:

acquire an acoustic signal generated during passage of a vehicle on a road;

convert the acquired acoustic signal into a frequency domain signal;

determine whether there is unexpectedness at each predetermined frequency using the frequency domain signal, wherein the unexpectedness is based at least on a short-time average level and a long-time average level of a spectrum of each frequency and a threshold set in advance; and

determine whether there is an abnormality in the road based on the number of frequencies at which it is determined that there is the unexpectedness.

2. An abnormality detection system comprising:

the abnormality detection device according to claim 1 and an abnormality location detection device,

wherein the processing circuitry of the abnormality detection device configured to further transmit positional information of the road on which it is determined that there is the abnormality,

the abnormality location detection device comprises processing circuitry configured to:

(i) receive the transmitted positional information,

(ii) generate a histogram by counting the number of positions corresponding to the received positional information in each pre-decided region, and

(iii) determine a location at which there is an abnormality based on the generated histogram.

3. The abnormality detection system according to claim 2 ,

wherein the processing circuitry of the abnormality location detection device is configured to further store a histogram, the histogram is generated previously by a histogram generation unit, and

the processing circuitry of the abnormality location detection device is further configured to:

determine a location at which there is an abnormality on the basis of the previously generated histogram read from the processing circuitry of the abnormality location detection device; and

determine that there is no abnormality with respect to the location where the abnormality was found previously when the location is presently determined that there is an abnormality on the basis of the generated histogram.

4. The abnormality detection system according to claim 2 ,

wherein the processing circuitry of the abnormality detection device is configured to further acquire traffic volume information of each location, and

the processing circuitry of the abnormality location detection device is further configured to:

calculate an abnormality detection ratio by dividing a count number included in the generated histogram by a traffic volume, and the traffic volume corresponding to the traffic volume information of a location corresponding to the count number; and

determine that there is an abnormality in a region in which the calculated abnormality detection ratio is greater than a predetermined threshold.

5. The abnormality detection device according to claim 1 ,

the processing circuitry of the abnormality detection device is further configured to:

acquire a vibration signal which is a signal of vibration generated during passage of a vehicle on a road; and

convert the acquired vibration signal into the frequency domain signal.

6. An abnormality detection method comprising:

acquiring an acoustic signal generated during passage of a vehicle on a road by a vibration acquisition unit;

converting the acquired acoustic signal into a frequency domain signal by a frequency domain conversion unit;

determining whether there is unexpectedness at each predetermined frequency using the frequency domain signal by an unexpectedness determination unit, wherein the unexpectedness is based at least on a short-time average level and a long-time average level of a spectrum of each frequency and a threshold set in advance; and

determining whether there is an abnormality in the road based on the number of frequencies at which it is determined that there is the unexpectedness by an abnormality determination unit.

7. A non-transitory computer readable medium that stores a program causing a computer to perform each step of the abnormality detection method according to claim 6 .