Shape recognition device, shape recognition program, and shape recognition method

A shape recognition system for recognizing a palm or a back of a hand, comprising:

• a memory to store computer executable instructions, and
• a processor to execute the computer executable instructions to perform processes, comprising:
• an outer shape detection process of detecting an outer shape of the hand;
• an extraction point setting process of setting a plurality of points inside of the detected outer shape of the hand as extraction points, target points are obtained by respectively superposing the outer shape of the hand onto the hand and projecting the extraction points onto a surface of the hand in the overlapping direction;
• a depth level detection process of measuring respective spatial distances from an infrared ray detection unit to the target points on the surface of the hand as depth levels, the target points respectively corresponding to the plurality of extraction points;
• and
• a hand orientation recognition process of determining which of a palmar side and a back side the hand shows, based on a criterion for fluctuations in the measured depth levels, the shape recognition program further comprising
• a reference point extraction process of extracting, from the detected outer shape, a central point of a maximum inscribed circle of the outer shape as a reference point, wherein
• the extraction point setting process includes setting a chord of the maximum inscribed circle such that the chord passes through the reference point and setting the plurality of extraction points at predetermined intervals onto the set chord.

A shape recognition method for recognizing a palm or a back of a hand comprising:

• an outer shape detection step of detecting an outer shape of the hand;
• an extraction point setting step of setting a plurality of points inside of the detected outer shape of the hand as extraction points, target points are obtained by respectively superposing the outer shape of the hand onto the hand and projecting the extraction points onto a surface of the hand in the overlapping direction;
• a depth level detection step of measuring respective spatial distances from an infrared ray detection unit to the target points on the surface of the hand as depth levels, the target points respectively corresponding to the plurality of extraction points;
• and
• a hand orientation recognition step of determining which of a palmar side and a back side the hand shows, based on a criterion for fluctuations in the measured depth levels, the shape recognition method further comprising
• a reference point extraction step of extracting, from the detected outer shape, a central point of a maximum inscribed circle of the outer shape as a reference point, wherein
• the extraction point setting step includes setting a chord of the maximum inscribed circle such that the chord passes through the reference point and setting the plurality of extraction points at predetermined intervals onto the set chord.