Method and system for detecting a false error on a component of a board inspected by an AOI machine

A data processing system, comprising:

• a processor;
• and
• an accessible memory, the data processing system particularly configured to:
• receive input data;
• wherein the input data comprise data originated from AOI machine's inspection results of a given inspected board marked as failed;
• apply a false error detector to the input data;
• wherein said detector is modeled with a trained function via an ML algorithm and wherein said detector generates output data, the trained function being trained with input training data and output training data, the input training data including data originating from the AOI machine's inspection results on components of a given set of AOI machine inspected faulty boards and the output training data including data originating from human inspection assessments on a subset of component errors of the given set that determine whether the component errors are false or true errors;
• provide output data;
• wherein the output data determines whether at least one of the component errors reported by the AOI machine for the given board is a false error.

A non-transitory computer-readable medium encoded with executable instructions that, when executed, cause one or more data processing systems to:

• receive input data, the input data comprising data originating from inspection results by an automated optical inspection (AOI) machine of a given inspected board that is marked as failed;
• apply a false error detector to the input data, the false error detector being modeled with a trained function via an ML algorithm and being configured to generate output data, via a machine learning (ML) algorithm and the detector generating output data, the trained function being trained with input training data and output training data, the input training data including data originating from the AOI machine's inspection results on components of a given set of AOI machine inspected faulty boards and the output training data including data originating from human inspection assessments on a subset of component errors of the given set that determine whether the component errors are false or true errors;
• and
• provide output data, the output data determining whether at least one component error reported by the AOI machine for the given board is a false error.

A method for providing, by a data processing system, a trained function for detecting one or more false errors among a set of errors detected on components of a board inspected by an automated optical inspection (AOI) machine, the method comprising:

• receiving input training data, the input training data comprising data originating from inspection results by an AOI machine on components of a given set of AOI machine-inspected faulty boards;
• receiving output training data, the output training data comprising data originating from a human inspection assessment on a subset of the component errors determining whether the errors are false or true errors;
• wherein the output training data is related to the input training data;
• training a function based on the input training data and the output training data via a machine learning (ML) algorithm;
• and
• providing the trained function for modeling a false error detector.

A method for detecting, by a data processing system, one or more false errors among a set of errors detected on components of a board inspected by an automated optical inspection (AOI) machine, the method comprising:

• receiving input training data, the input training data comprising data originating from inspection results of an AOI machine inspection on components of a given set of AOI machine-inspected faulty boards;
• receiving output training data, the output training data comprising data originating from human inspection assessments on a subset of the component errors determining whether the errors are false or true errors;
• wherein the output training data is related to the input training data;
• training a function based on the input training data and the output training data via a machine learning (ML) algorithm, to form a trained function;
• providing the trained function for modeling a false error detector;
• receiving input data, the input data comprising data originating from the inspection results of the AOI machine inspection of a specific inspected board marked as failed;
• applying the false error detector to the input data;
• wherein the detector is modeled with the trained function and wherein the detector generates output data;
• and
• providing the output data which determine whether at least one of the component errors reported by the AOI machine for the specific inspected board is a false error.