“…In order to classify the bearings into defect-free or defective class (Figure 4), the approach is proposed that is used to control products by the method of free oscillations [11][12][13][14] characteristic of the procedures for anomaly rejection: the analysis algorithm interprets a set (Table 1) for defect-free (bearing number 1-2) and defective rolling bearings (3)(4)(5)(6)(7)(8)(9)(10)(11)(12)(13)(14) with the amplitude of the noise component below a) and above b) defect frequency amplitude. The comparison of signal scale graphs by PSNR metrics was performed based on forty types of wavelet basis, such as Meyer, Morlet, MexicanHat, Haar, db02, db03, db04, db05, db06, db07, db08, db09, db10, db11, db12, db13, db14, sym2, sym3, sym4, sym5, sym6, sym7, sym8, coif1, coif2, coif3, coif4, coif5, bior1_3, bior1_5, bior2_2, bior2_4, bior2_6, bior2_8, bior3_1, bior3_3, bior3_5, bior3_7, bior3_9 (Table 2) [12,13].…”