Image processing apparatus, image processing method, and non-transitory computer readable medium
An image processing apparatus includes a first acquiring unit that acquires an image to be processed; a setting unit that sets multiple partial image areas in the image to be processed; a second acquiring unit that acquires a first classification result indicating a possibility that an object of a specific kind is included in each of the multiple partial image areas; and a generating unit that generates a second classification result indicating a possibility that the object of the specific kind is included in the image to be processed on the basis of the first classification result of each of the multiple partial image areas.
1. An image processing apparatus comprising:
a first acquiring unit that acquires an image to be processed;
a setting unit that sets a plurality of partial image areas in the image to be processed;
a second acquiring unit that, for each of the plurality of partial image areas, acquires a first classification result indicating a possibility that an object of a specific kind is included in the partial image area; and
a generating unit that generates a second classification result indicating a possibility that the object of the specific kind is included in the image to be processed on the basis of the first classification results of the plurality of partial image areas,
wherein the first classification result is a probability that the object of the specific kind is included, and
wherein the generating unit generates the second classification result indicating the possibility that the object of the specific kind is included in the image to be processed on the basis of a sum or a product of the probabilities, which are the first classification results of the respective plurality of partial image areas, or a result of comparison between a number of the first classification results having the probabilities higher than or equal to a threshold value and a number of the first classification results having the probabilities lower than the threshold value.
2. The image processing apparatus according to claim 1 , further comprising:
an extracting unit that extracts part of the first classification result from the first classification results of the respective plurality of partial image areas,
wherein the generating unit generates the second classification result indicating the possibility that the object of the specific kind is included in the image to be processed on the basis of the first classification result extracted by the extracting unit.
3. The image processing apparatus according to claim 2 ,
wherein the first classification result is a probability that the object of the specific kind is included, and
wherein the extracting unit extracts the first classification results of a predetermined number or ratio counted from the first classification result having a highest probability, among the first classification results of the respective plurality of partial image areas.
4. The image processing apparatus according to claim 2 ,
wherein the first classification result is a probability that the object of the specific kind is included, and
wherein the extracting unit extracts the first classification results having the probabilities higher than or equal to a threshold value, among the first classification results of the respective plurality of partial image areas.
5. The image processing apparatus according to claim 1 , further comprising:
a unit that causes a classifier to learn a classification condition of the object of the specific kind on the basis of an image feature of each of the partial image areas set for each of one or more sample images including the object of the specific kind,
wherein the second acquiring unit acquires the first classification result indicating the possibility that the object of the specific kind is included in each of the plurality of partial image areas with the classifier on the basis of the image feature of the partial image area.
6. The image processing apparatus according to claim 5 , further comprising:
a partial area information learning unit that learns a feature concerning a position and a size of each of the partial image areas set in the sample image,
wherein the setting unit sets the plurality of partial image areas in the image to be processed on the basis of the feature concerning the position and the size of each of the partial image areas learned by the partial area information learning unit.
7. The image processing apparatus according to claim 6 ,
wherein the image feature of each of the partial image areas is generated on the basis of a distribution of a local feature of each of one or more pixels included in the partial image area.
8. The image processing apparatus according to claim 6 ,
wherein the image feature of each of the partial image areas is generated on the basis of a distribution of a local feature of each of one or more pixels included in the partial image area.
9. The image processing apparatus according to claim 1 ,
wherein the setting unit sets the plurality of partial image areas so that at least some of the plurality of partial image areas set in the image to be processed are overlapped with each other.
10. A non-transitory computer readable medium storing a program causing a computer to execute a process comprising:
acquiring an image to be processed;
setting a plurality of partial image areas in the image to be processed;
acquiring, for each of the plurality of partial image areas, a first classification result indicating a possibility that an object of a specific kind is included in the partial image area; and
generating a second classification result indicating a possibility that the object of the specific kind is included in the image to be processed on the basis of the first classification result of the plurality of partial image areas,
wherein the first classification result is a probability that the object of the specific kind is included, and
wherein the generating unit generates the second classification result indicating the possibility that the object of the specific kind is included in the image to be processed on the basis of a sum or a product of the probabilities, which are the first classification results of the respective plurality of partial image areas, or a result of comparison between a number of the first classification results having the probabilities higher than or equal to a threshold value and a number of the first classification results having the probabilities lower than the threshold value.
11. An image processing method comprising:
acquiring an image to be processed;
setting a plurality of partial image areas in the image to be processed;
acquiring, for each of the plurality of partial image areas, a first classification result indicating a possibility that an object of a specific kind is included in the partial image area; and
generating a second classification result indicating a possibility that the object of the specific kind is included in the image to be processed on the basis of the first classification result of each of the plurality of partial image areas,
wherein the first classification result is a probability that the object of the specific kind is included, and
wherein the generating unit generates the second classification result indicating the possibility that the object of the specific kind is included in the image to be processed on the basis of a sum or a product of the probabilities, which are the first classification results of the respective plurality of partial image areas, or a result of comparison between a number of the first classification results having the probabilities higher than or equal to a threshold value and a number of the first classification results having the probabilities lower than the threshold value.