Printed circuit board inspecting apparatus, method for detecting anomaly in solder paste and computer readable recording medium

A non-transitory computer-readable recording medium having a program to be executed by a computer, wherein the program includes executable instructions that instruct a processor, when the program is executed by the processor, to perform:

• obtaining measurement shape information about each of a plurality of solder pastes printed on a first printed circuit board through a plurality of apertures formed in a stencil and aperture shape information about each of the plurality of apertures;
• obtaining probability values that a first solder paste printed through a first aperture of the plurality of apertures and each of a plurality of second solder pastes printed through second apertures other than the first aperture of the plurality of apertures have the measurement shape information when the first solder paste and the plurality of second solder pastes are printed on the first printed circuit board, by applying the obtained measurement shape information and the obtained aperture shape information to a machine-learning based model, and
• detecting whether an anomaly in the first solder paste occurred based on the obtained probability values, and
• wherein the machine-learning based model is derive a probability value that each of two solder pastes have measurement shape information about each of the two solder pastes when the two solder pastes are printed on a printed circuit board through two apertures having a predetermined aperture shape, which are formed in a stencil, using aperture shape information of the two apertures and the measurement shape information about each of the two solder pastes.

A method for detecting an anomaly in a solder paste by a printed circuit board inspection apparatus, the method comprising:

• obtaining measurement shape information about each of a plurality of solder pastes printed on a first printed circuit board through a plurality of apertures formed in a stencil and aperture shape information about each of the plurality of apertures;
• obtaining probability values that a first solder paste printed through a first aperture of the plurality of apertures and each of a plurality of second solder pastes printed through second apertures other than the first aperture of the plurality of apertures have the measurement shape information when the first solder paste and the plurality of second solder pastes are printed on the first printed circuit board, by applying the obtained measurement shape information and the obtained aperture shape information to a machine-learning based model, and
• detecting whether an anomaly in the first solder paste occurred based on the obtained probability values, and
• wherein the machine-learning based model is derive a probability value that each of two solder pastes have measurement shape information about each of the two solder pastes when the two solder pastes are printed on a printed circuit board through two apertures having a predetermined aperture shape, which are formed in a stencil, using aperture shape information of the two apertures and the measurement shape information about each of the two solder pastes.