Abstract: Provided is a visual inspection device including: an image pickup unit configured to pick up an image of a surface of an object to be inspected; a posture control unit configured to change a posture of the object to be inspected with respect to the image pickup unit; a storage unit having stored therein a learning result obtained by machine learning using a defective product sample image, the defective product sample image being generated by combining a defect portion image prepared in advance with a non-defective product image of the object to be inspected at a combination position obtained by referring to three-dimensional data corresponding to the object to be inspected tilted so as to form a posture angle set in advance; and an inspection unit configured to inspect the surface of the object to be inspected based on a picked-up image picked up at the posture angle by the image pickup unit and on the learning result.

Abstract: A visual inspection apparatus detects a defect candidate for each of captured images of a crown surface of a piston that are captured at a plurality of posture angles different from one another, locates a three-dimensional position of the defect candidate based on a two-dimensional position of the defect candidate detected from at least one captured image among the captured images captured at the plurality of posture angles, applies a perspective projection transformation on the three-dimensional position of the defect candidate to acquire a two-dimensional position of the defect candidate for each of the captured images captured at the plurality of posture angles, determines a feature amount regarding the defect candidate based on the two-dimensional position of the defect candidate for each of the captured images captured at the plurality of posture angles, and inspects the crown surface of the piston using the feature amount regarding the defect candidate.

Abstract: A visual inspection apparatus inspects a surface of a piston based on a captured image acquired by a camera and a learning result acquired by conducting machine learning using a plurality of defective product sample images, which is each generated by combining a two-dimensional image of a defect image that is generated based on a three-dimensionally pre-generated defect model with an image of the surface of the piston.

Abstract: A visual inspection apparatus of the present invention includes a rotational portion configured to rotate in a state that an inspection target is placed on the rotational portion, a mirror portion provided so as to reflect the inspection target in the state that the inspection target is placed on the rotational portion, an imaging portion configured to image the inspection target reflected in the mirror portion, an adjustment portion configured to change an imaging angle of the imaging portion relative to the inspection target reflected in the mirror portion, and an inspection portion configured to inspect a surface of the inspection target based on image data captured by the imaging portion.