In one embodiment, a system for inspecting an object comprises a first camera for inspecting a first surface of the object, and a second camera for inspecting a second surface of the object. The object may be placed upon a support structure during simultaneous inspection by the first camera and the second camera. At least one roller is arranged to selectively engage the object when the object is placed upon the support structure, wherein the at least one roller is adapted for circumferential rotation relative to the support structure. Rotation of the at least one roller causes a corresponding circumferential rotation of the object relative to the first and second cameras.

In one embodiment, a system for inspecting an object comprises a first camera for inspecting a first surface of the object, and a second camera for inspecting a second surface of the object. The object may be placed upon a support structure during simultaneous inspection by the first camera and the second camera. At least one roller is arranged to selectively engage the object when the object is placed upon the support structure, wherein the at least one roller is adapted for circumferential rotation relative to the support structure. Rotation of the at least one roller causes a corresponding circumferential rotation of the object relative to the first and second cameras.

An artificial neural network-based method for detecting a surface type of an object includes: receiving a plurality of object images, wherein a plurality of spectra of the plurality of object images are different from one another and each of the object images has one of the spectra; transforming each object image into a matrix, wherein the matrix has a channel value that represents the spectrum of the corresponding object image; and executing a deep learning program by using the matrices to build a predictive model for identifying a target surface type of the object. Accordingly, the speed of identifying the target surface type of the object is increased, further improving the product yield of the object.

An artificial neural network-based method for detecting a surface type of an object includes: receiving a plurality of object images, wherein a plurality of spectra of the plurality of object images are different from one another and each of the object images has one of the spectra; transforming each object image into a matrix, wherein the matrix has a channel value that represents the spectrum of the corresponding object image; and executing a deep learning program by using the matrices to build a predictive model for identifying a target surface type of the object. Accordingly, the speed of identifying the target surface type of the object is increased, further improving the product yield of the object.

A device for automatic control of color tone of a reel of thread includes a frame supporting an arm on which a reel of thread is loadable, a meter configured to project a measuring beam onto a cylindrical portion of the reel of thread, the meter having a camera and an illuminator aligned with the camera, the illuminator being multispectral and configured to illuminate the reel of thread with different wavelengths. A computer provided with a screen with graphic interface and connected to the meter processes the classification of the reel of thread loaded on the arm on the basis of measurements performed by the meter.

An inspection apparatus (100) detects an inspection object (90) from first image data (10) in which the inspection object (90) is included. The inspection apparatus (100) generates second image data (20) by performing a geometric transform on the first image data (10) in such a way that a view of the detected inspection object (90) becomes a view satisfying a predetermined reference. In an inference phase, the inspection apparatus (100) detects, by using an identification model for detecting an abnormality of the inspection object (90), an abnormality of the inspection object (90) included in the second image data (20). Further, in a learning phase, the inspection apparatus (100) learns, by using the second image data (20), an identification model for detecting an abnormality of the inspection object (90).

This coating condition detection method according to one embodiment uses a simple device structure to detect the coating condition of a resin layer of a coated fiber. Under the coating condition detection method, an imaging optical system including a reflection mirror having a guide hole through which the optical fiber passes is prepared, and the imaging optical system is disposed so as to cause an object plane conjugate with an imaging plane to intersect the optical fiber that has passed through the reflection mirror and forms an image of light released from the optical fiber on the imaging plane to detect intensity of light at each point on the imaging plane with the intensity of light associated with information on a corresponding position on the object plane.

This coating condition detection method according to one embodiment uses a simple device structure to detect the coating condition of a resin layer of a coated fiber. Under the coating condition detection method, an imaging optical system including a reflection mirror having a guide hole through which the optical fiber passes is prepared, and the imaging optical system is disposed so as to cause an object plane conjugate with an imaging plane to intersect the optical fiber that has passed through the reflection mirror and forms an image of light released from the optical fiber on the imaging plane to detect intensity of light at each point on the imaging plane with the intensity of light associated with information on a corresponding position on the object plane.

An artificial neural network-based method for selecting a surface type of an object includes receiving at least one object image, performing surface type identification on each of the at least one object image by using a first predictive model to categorize the object image to one of a first normal group and a first abnormal group, and performing surface type identification on each output image in the first normal group by using a second predictive model to categorize the output image to one of a second normal group and a second abnormal group.

An artificial neural network-based method for selecting a surface type of an object includes receiving at least one object image, performing surface type identification on each of the at least one object image by using a first predictive model to categorize the object image to one of a first normal group and a first abnormal group, and performing surface type identification on each output image in the first normal group by using a second predictive model to categorize the output image to one of a second normal group and a second abnormal group.