Wood Color Image Enhancement Recognition Based On DWT-SVD And Inflection Gamma

Wood recognition is one of the important research directions in the field of wood.People usually recognize wood by visual analysis.However,this method is not only time-consuming and laborious,but also ineffective,so computer recognition is particularly important.Influenced by various factors in the process of image acquisition,the image quality is poor,and people can not get the desired information,which affects the wood recognition rate.In order to enhance dark color image and improve wood recognition rate,an improved inflection point transfer function and gamma correction color image enhancement algorithm based on discrete wavelet transform(DWT)and singular value decomposition(SVD)is proposed in this paper.By using DWT,the input image is decomposed into four frequency sub-bands,and the singular value matrix of low frequency sub-band image is calculated.The horse correction is used to calculate the inflection point transfer function to further improve the LL component.After that,the processed LL-band image,along with the unprocessed LH,HL and HH subbands,undergoes inverse wavelet decomposition to generate enhanced images,which are applied to the wood field.After enhancement,the support Vector Machine(SVM)radial basis function was used to identify poplar wood chips and bark,and the micro-cross sections of Pinus sylvestris and Poplar were identified.The main conclusions of the paper are as follows:(1)This method is compared with four traditional enhancement methods:histogram equalization,histogram specification,wavelet singular value decomposition and inflection point correction.It not only enhances the contrast and brightness of the image,but also improves the quality of the image.Subjective analysis shows that this method has the best enhancement effect.In objective analysis,PSNR(Peak Signal-to-Noise Ratio),Mean Square Transform,SSIM(Structural Similarity Index Measurement)and FSIM(Feature Similarity Index Measurement)are the four evaluation indexes of PSNR(Peak Signal-to-Noise Ratio),Mean Square Transform(MSE),Structural Similarity Index Measurement(SSIM),and Feature Similarity Index Measurement(FSIM).Improvement in varying degrees.(2)The macro-image of Poplar wood chips and bark is used as the original image for recognition.First,image enhancement preprocessing is carried out.Then,SVM radial basis function is used to recognize Poplar wood chips and bark.The recognition effect of this method is obviously better than that of traditional methods,and the recognition rate has been significantly improved.The recognition rate of Poplar wood slices and Poplar bark of identification parameter angle MS was increased by 0.7%and 0.30%,the recognition rate of Poplar wood slices and Poplar bark of identification parameter brightness MSE was increased by 1.2%and 0.9%,the recognition rate of Poplar wood slices and Poplar bark of identification parameter angle x component was increased by 0.8%and 0.5%,and the identification parameter angle Y component was increased by 0.4%and Poplar bark by 0.6%.(3)Taking the micro-images of the cross-section of Pinus sylvestris and Poplar as the original image,image enhancement pretreatment is carried out first,and then the micro-cross-section of Pinus sylvestris and Poplar are recognized by using SVM radial basis function.The recognition effect of this method is obviously better than that of traditional methods,and the recognition rate has been improved obviously.The microcosmic cross section of Pinus sylvestris and Poplar increased by 0.7%and 0.5%respectively,the microcosmic cross section of Pinus sylvestris and poplar increased by 0.5%and 0.2%respectively,the microcosmic cross section of Pinus sylvestris and Poplar increased by 0.4%and 0.6%respectively,and the microcosmic cross section of Poplar increased by 0.7%and 0.5%respectively.The cross-section of Poplar is increased by 0.6%,the micro-cross-section of Poplar is increased by 0.9%.