Impression evaluation of an image based on similarity to a reference image

1 . An impression analysis system, comprising:

an input obtainer that receives an evaluation target image; and

a hardware processor,

wherein

the hardware processor:

extracts a low-order image feature amount, and a high-order image feature amount, from the evaluation target image, and

compares data obtained by extracting a low-order image feature amount and a high-order image feature amount from a reference image associated with an impression evaluation, with data of the image feature amounts extracted by the hardware processor, and calculates a similarity between the evaluation target image and the reference image,

wherein the low-order image feature amount includes at least any of a color, a luminance distribution, a bearing, contrast, a face, a font, and a motion, and

wherein the high-order image feature includes at least any of a degree of position bias and a processing fluency feature, wherein the processing fluency feature includes a degree of complexity, a design density, and aspatial frequency.

2 . The impression analysis system according to claim 1 , wherein the hardware processor proposes an improvement about an impression of the evaluation target image, in accordance with a value of the extracted image feature amount.

3 . The impression analysis system according to claim 1 , wherein the degree of complexity is a local fractal dimension.

4 . The impression analysis system according to claim 1 , wherein the reference image is freely selected by an evaluator.

5 . The impression analysis system according to claim 1 , wherein the hardware processor:

calculates a coloration pattern ratio in the evaluation target image;

extracts a saliency as the high-order image feature amount; and

assigns a weight based on the extracted saliency in calculating the coloration pattern ratio.

6 . The impression analysis system according to claim 5 , wherein the hardware processor excludes a transparent area residing in the evaluation target image in calculating the coloration pattern ratio.

7 . The impression analysis system according to claim 5 , wherein the hardware processor assigns a predetermined gradation to a transparent area residing in the evaluation target image, and extracts the saliency.

8 . An impression analysis method, comprising:

extracting a low-order image feature amount, and a high-order image feature amount, from an input evaluation target image; and

similarity determining that compares data obtained by extracting a low-order image feature amount and a high-order image feature amount from reference image associated with an impression evaluation, with data of the image feature amounts extracted by the feature amount extracting, and calculates a similarity between the evaluation target image and the reference image,

wherein the low-order image feature amount includes at least any of a color, a luminance distribution, a bearing, contrast, a face, a font, and a motion, and

wherein the high-order image feature includes at least any of a degree of position bias and a processing fluency feature, wherein the processing fluency feature includes a degree of complexity, a design density, and aspatial frequency.

9 . A non-transitory computer readable recording medium storing a program causing a computer to perform:

extracting a low-order image feature amount, and a high-order image feature amount, from an input evaluation target image; and

similarity determining that compares data obtained by extracting a low-order image feature amount and a high-order image feature amount from reference image associated with an impression evaluation, with data of the image feature amounts extracted by the feature amount extracting, and calculates a similarity between the evaluation target image and the reference image,

wherein the low-order image feature amount includes at least any of a color, a luminance distribution, a bearing, contrast, a face, a font, and a motion, and

wherein the high-order image feature includes at least any of a degree of position bias and a processing fluency feature, wherein the processing fluency feature includes a degree of complexity, a design density, and aspatial frequency.

10 . The impression analysis system according to claim 1 , wherein the high-order image feature includes a degree of position bias of where the line of sight tends to be concentrated.