Russian Federation
Russian Federation
The increasing complexity of integrated circuits and stricter reliability requirements have led to the need to improve the efficiency of automated quality control at the stages of microcircuit assembly and testing [1; 5]. Standard automatic optical inspection (AOI) systems based on comparison with a reference image provide a high defect detection rate, but in real-world production they generate a significant number of false alarms, overloading operators and reducing line productivity [2; 6]. In recent years, convolutional neural network (CNN) methods have been actively implemented to detect defects in printed circuit boards, microcircuit packages, and internal structural faults, including using scanning acoustic tomography data [3; 8; 10]. The objective of this work is to analyze current research to substantiate approaches to improving the efficiency of automated microcircuit quality control (AQC) by combining AQC and machine learning methods without radically replacing existing measurement equipment [1–3; 5–8]. It is demonstrated that the use of retrainable CNN classifiers, thoughtful generation of training samples, and the integration of optical and acoustic inspection can reduce the false alarm rate of AQC systems severalfold while maintaining or increasing sensitivity to real defects [2–6; 8–10].
AQC, AQC, scanning acoustic tomography, convolutional neural networks, false alarms
1. Vlasov, A. I. Koncepciya universal'nogo intellektual'nogo smart-avtomata / A. I. Vlasov, S. S. Filin, A. I. Krivoshein // Prom-inzhiniring : Trudy Vserossiyskoy nauchno-tehnicheskoy konferencii. – Chelyabinsk, 2019. – S. 267-271.
2. Rubcov, Yu. V. Avtomaticheskiy vizual'nyy kontrol' kachestva izdeliy mikroelektroniki metodom etalonnyh shablonov / Yu. V. Rubcov, V. E. Malyshev, A. A. Nazarenko // Radioelektronnye prilozheniya: problemy i ih resheniya. – 2024. – № 13. – S. 18-21.
3. Usmanov, A. I. Analiz sovremennyh metodov avtomatizacii nerazrushayuschego kontrolya kachestva i diagnostiki elektronnyh komponentov / A. I. Usmanov, A. V. Kozlova, V. G. Meshkov // Vestnik MGTU «STANKIN». – 2024. – № 2 (69). – S. 161-172.
4. Bykov, A. E. Opredelenie pravil'nogo raspolozheniya kontrol'nyh operaciy v tehnologicheskom processe izgotovleniya mikroshem / A. E. Bykov, E. I. Lazareva // Social'no-ekonomicheskie i tehnicheskie problemy oboronno-promyshlennogo kompleksa Rossii: istoriya, real'nost', innovacii : mezhvuzovskiy sbornik statey po materialam VII Vserossiyskoy nauchno-prakticheskoy konferencii ; Nizhegorodskiy gosudarstvennyy tehnicheskiy universitet im. R.E. Alekseeva. – Nizhniy Novgorod, 2021. – S. 152-155.
5. Avakov, S. M. Principial'noe proektirovanie i razrabotka programmnogo obespecheniya dlya obespecheniya kontrolya kachestva topologicheskih struktur v mikroelektronike / S. M. Avakov, A. A. Voronov, V. V. Ganchenko // Vestnik Polockogo gosudarstvennogo universiteta. Seriya S. Fundamental'nye nauki. – 2024. – № 2 (43). – S. 2-9.
