Chinese Version Light学术出版中心
Latest Issue
Issue 1 ， 2021
-
Large field of view line-scanning confocal holographic microscopy
Abstract：Traditional microscopic imaging generally records the intensity information of a sample， but the contrast is poor for translucent or phase tissue. To realize phase tissue non-fluorescence labeling imaging， a line-scanning confocal holographic imaging method was adopted. In line-scanning confocal imaging， a reference light was added to form off-axis image plane digital holography at the confocal slit. The sample was scanned by controlling the movement of the sample. The interference lines obtained were combined to form a two-dimensional hologram， and the distribution of amplitude and phase was obtained by filtering in the frequency domain. The phase striation caused by environmental vibration was rectified using the characteristics of adjacent sections， and large field-of-view holographic imaging was realized by multi-area scanning and splicing. A USAF1951 resolution plate was imaged by line scanning confocal holography， and a jitter correction algorithm was used to reduce the jitter stripes in the reconstructed phase map by 84.7% to obtain three sub-area maps. An imaging field of 1 160 μm × 1 043 μm was achieved by splicing； scanning more sub-regions can help in obtaining a larger field of view and in realizing confocal phase imaging of onion epidermal cells. The experimental results show that the line-scanning confocal holographic imaging method can realize large-field phase imaging of translucent samples and provide guidance and a basis for the development of related instruments.Keywords：digital holography;optical microscopy;line-scanning confocal;scanning splicing;large field of view- 1242
- |
- 156
- |
- 3
Published：2021-02-04 -
Experimental characteristics and measurement accuracy evaluation of space cryogenic absolute radiometric primary benchmark
Abstract：The establishment of a high-precision onboard radiometric benchmark is the primary technical means to improve on-orbit radiometric calibration accuracy. The space cryogenic absolute radiometric benchmark traced to the SI has been researched. First， a high-sensitivity cryogenic absolute radiation detector was developed. A high-thermal-resistance heat transfer structure was designed to connect the detector and cold finger to improve sensitivity. A highly stable thermal environment， better than 0.4 mK， was established by multistage precise temperature control. Then， a two-stage pulse tube cooler with a large cooling capacity was used to obtain a working environment of 20 K through a Stirling cycle in which the working fluid gas was compressed. Second， an absolute radiation measurement link was built to evaluate the measurement uncertainty. Finally， an indirect comparison was carried out using the reference cryogenic radiometer of the National Institute of Metrology of China. The experimental results illustrate that the sensitivity of the space cryogenic absolute radiometer experimental prototype was 3 565 K/W， the repeatability of 0.4 mW laser power measurement was 0.017%， and the relative standard uncertainty was 0.029%. The normalized deviation of the benchmark cryogenic radiometer was 0.4， and the validity of the uncertainty evaluation results was verified. The research lays a key technical foundation for the development of space cryogenic absolute radiometers. It is important to examine on-orbit radiometric calibration methods based on absolute detectors.Keywords：radiometric calibration;cryogenic absolute radiometer;electrical substitution;uncertainty;indirect comparison- 358
- |
- 42
- |
- 2
Published：2021-02-04 -
Preparation and characterization of ScAlN thin film with high scandium concentration
Abstract：In this study， a novel seed layer structure was used to prepare ScAlN piezoelectric thin films through pulsed DC reactive magnetron sputtering at room temperature with scandium aluminum alloy as the target source after optimizing the process parameters. A novel structure for d33 measurement was designed and fabricated to solve the problem that the wet etch process is still not mature. The crystal quality， components， and electromechanical coupling factor were measured using X-ray diffraction， scanning electron microscopy， energy-dispersive X-ray spectroscopy， quasi-static d33 analysis， and nanoindentation. The Sc0.35Al0.65N film prepared was highly c-axis oriented， with a full width at half maximum of 2.167°. In addition， the film had a high crystal density with a few shell-like convex parts. The test results show that the piezoelectric constant d33 of the film is -23.4 pC/N， and the electromechanical coupling factors- 344
- |
- 29
- |
- 0
Published：2021-02-04 -
Development of optical waveguide elements in augmented reality near-eye displays
Abstract：Augmented reality （AR） technology is a new type of display technology that superimposes information from a virtual environment onto the real world and makes effective use of it. It has broad application prospects in education， medical care， tourism， automobiles， construction， and other fields. AR near-eye display devices are a major part of AR technology， and optical waveguides are the core optical elements used to realize high performance and miniaturization of AR near-eye displays. Optical waveguides are primarily divided into geometrical and diffractive optical waveguides. The principle behind geometrical optical waveguides is simple. Its preparation technology is relatively complete and can be mass produced. However， the field of view and exit pupil range are small， the preparation process is complicated， and the diffractive optical waveguides rely on an extremely high degree of design freedom to obtain excellent performance. In addition， the problems of rainbow effect and dispersion in imaging must be solved. This study mainly focuses on the design and manufacturing technology of different types of optical waveguides. It also analyzes the advantages and disadvantages of geometrical and diffractive optical waveguides in terms of principles， structural parameters， performance evaluation， and preparation process. Finally， the study summarizes the challenges faced by current AR near-eye display technology and looks forward to future developmental prospects.Keywords：augmented reality;geometrical optical waveguide;diffractive optical waveguide;fabrication technology;micro-machining;nanoimprint- 648
- |
- 132
- |
- 4
Published：2021-02-04 -
Optical design of echelle spectrometer with ultra-wide wavelength range and high resolution
Abstract：In order to realize the echelle spectrometer system with ultra-wide wavelength range and high resolution， an optical path structure design was proposed， and the detection method of deep ultraviolet lights was studied and verified. Frist， the optical structure combined the advantages of quasi-littrow structure and C-T structure，so as to ensure the dispersing path has the characteristics of high diffraction efficiency and the stray light was well suppressed. In the limited range of optional materials，the multi-optimization method was employed to design the optical parameters. In order to correct the aberration and improve the spectral resolution， a correction structure composed of spherical and cylindrical lenses was added. Finally， a solution to the detection of deep ultraviolet lights was proposed， and the feasibility of the scheme was verified by finite element analysis. The final design results indicates that in the wavelength range of 160-1 000 nm， the imaging spot RMS value is better than 12.1 µm， the spectral resolution of 257.61 nm is better than 0.009 nm. It can meet the dispersion demand of ultra-wide wavelength range and high resolution detection system.Keywords：optical design;echelle spectrometer;ultra-wide wavelength range;high resolution- 304
- |
- 59
- |
- 2
Published：2021-02-04 -
Calibration of line-structure light plane using non-specific target
Abstract：High-precision calibration of a line-structure light plane can not be achieved if the measurement systems lack a target plane. To address this problem， this paper proposes a line-structure light plane calibration method using non-specific targets； this method is based on the principle of stereo vision and the line-structure light measurement system. First， a non-specialized target is placed on the line-structure light plane， and point clouds formed by projecting line-structure light on the target are acquired using the parallax method. Multiple point clouds are obtained by varying the posture of the non-specific target. The least squares method is then used to perform plane fitting on these collected multiple point cloud sets， and subsequently， the structure light plane parameters are determined to complete plane calibration. A human arm and 304 steel plate were adopted as the targets， and the proposed method was used for the calibration of the line-structure light plane， where these targets were located. Using this calibrated line-structure light plane， the dimensions of five square workpieces were measured. The results indicate that the average measurement accuracy of the proposed calibration method is approximately ±0.056 mm. This proves that the method can be used for calibration， without requiring a special target， while also meeting the accuracy requirements of the measurement system.Keywords：precision measurement;Line structured light;light plane calibration;non-specific targets- 240
- |
- 27
- |
- 0
Published：2021-02-04 -
Spatial resolution measurement of industrial CT system based on equal diameter circumscribed circle phantom
Abstract：Spatial resolution measurement of industrial CT based on a periodic structure phantom has strict requirements on the machining accuracy of the phantom， which affects the practicability and accuracy of the method. To address this issue， a spatial resolution measurement technique of industrial CT based on an equal-diameter circumscribed circle is proposed. First， the measurement mechanism of the line-pair-card is analyzed. The equal-diameter circumscribed circle phantom was designed to form a continuous variable-spacing structure. The equivalence relation between the variable spacing structure and strip structure （line-pair-card） is derived. A spatial resolution measurement technique of industrial CT is established based on an equal-diameter circumscribed circle. Then， an automatic measurement method of spatial resolution is realized using an image processing algorithm. The results of this method are similar to those of the line-pair-card measurement， and the relative error is within 10%. This method has fewer requirements for the machining accuracy of phantoms. The measurement process is simple and practical.Keywords：industrial computerized tomography;spatial resolution;equal-diameter circumscribed circle;modulation transfer function- 187
- |
- 25
- |
- 0
Published：2021-02-04 -
Sensitivity theoretical analysis and desensitization design method for reflective optical system based on optical path variation
Abstract：The final performance of an optical system depends not only on its theoretical design， but more importantly， on various error-causing factors such as surface processing， alignment position， and system stability during its construction. To explore the influencing factors and the error sensitivity of a reflective optical system， a coaxial two-mirror system is considered as the research object. The optical path variation （OPV） caused by a misalignment error is proposed as the evaluation index of the error sensitivity of the optical system， and the mathematical expression of the OPV caused by the mirror position misalignment factor is deduced theoretically. By using a standard Ritchey-Chretien system and the ray tracing method， the OPV caused by various disturbances is calculated. Compared with the theoretical value， the relative error is approximately 1%， which proves the accuracy of the theoretical expression of the OPV. Based on the theoretical study of OPV， a desensitization design method for a coaxial reflective optical system with an OPV is proposed and established. Considering a coaxial two-mirror system with a focal length of 5 600 mm as an example， an optical system that meets the OPV and wavefront error index is designed after 15 rounds of iterative optimization. The accuracy of the OPV as the sensitivity evaluation standard and the effectiveness of the desensitization design method are verified by the relationship between the OPV and the wavefront error modification.Keywords：optical design;reflective optical system;sensitivity;desensitization design method;optical path variation- 451
- |
- 67
- |
- 1
Published：2021-02-04
Modern Applied Optics
-
Design of spiral flat transmitting coil pair for intestinal robot Cover Article
Abstract：Traditional wireless power transmission （WPT） systems used in intestinal robots generally adopt a hollow cylinder axial winding mode， which occupies a large amount of space， limits the activity space of the examinee， and complicates the winding. This paper proposes a transmitting coil structure for a loosely coupled WPT system of an intestinal robot. The proposed structure is composed of a square spiral wound flat coil pair， which substantially reduces the axial length of the coil. Thereby， it effectively reduces the volume of the transmitting coil and makes the coil structure simpler and lighter. First， the working principle of the WPT system is introduced. Then， the magnetic field distribution of the designed coil is analyzed via COMSOL simulation. Finally， the feasibility of the design of the power transmitting device is verified experimentally， and the optimal resonant frequency is determined. The experimental results show that when the working frequency is 213 kHz， the power of the WPT system at the center of the coil pair is 787 mW， and the transmission efficiency is 5.7%. Thus， the proposed structure can satisfy the power requirements of an intestinal robot.Keywords：Intestinal robot;Transmitting coil pair;wireless power transmission;resonant frequency- 211
- |
- 31
- |
- 3
Published：2021-02-04 -
Design of V-shaped linear ultrasonic motor driven with low-voltage
Abstract：To solve the problem of the high driving voltage of traditional V-shaped linear ultrasonic motors， which results in a complex driving circuit and the disadvantage of structure miniaturization when matching with transformer driving， a low voltage driving V-shaped linear ultrasonic motor based on laminated piezoelectric ceramics was proposed. Based on the analysis of the working principle of the V-shaped linear ultrasonic motor， a V-shaped oscillator with laminated piezoelectric ceramics and a clamping device were designed. Finally， a prototype of the motor was fabricated， and the impedance and performance were measured. The experimental results show that the operating frequency of the motor is between 36 and 38 kHz while under a driving voltage of 50 Vpp and 37 kHz， the maximum output force of the motor reaches 25.8 N， and the maximum no-load velocity reaches 1.221 m/s.Keywords：ultrasonic motor;linear motor;V-shaped structure;low voltage drive;vibrator design- 191
- |
- 25
- |
- 1
Published：2021-03-09 -
Vision measurement of pitch and profile deviations for small modulus gears with unknown parameters
Abstract：Using contact measurements for small modulus gears is difficult owing to the limited tooth spaces. This work focuses on the vision measurement of small modulus gears with unknown parameters. Based on the sub-pixel technique used in digital image processing， the gear measurement datum， which is the geometric center of the gear， is located. Thereafter， the tooth number， modulus， diameter of the tooth addendum circle， and diameter of the tooth dedendum circle are calculated. According to the definition of gear accuracy in ISO 1328-1：2013， this study introduces an evaluation method and a measurement process for gear pitch deviations and profile deviations based on visual measurements， and an evaluation software is developed. A visual measurement test on the pitch and profile deviations of involute cylindrical spur gears with a modulus of 0.5 mm was carried out， and the results were compared with those from a gear measurement center. The comparison indicated that the absolute error in the left and right tooth surface measurements had a maximum value of 4 μm and a minimum value of 1 μm； the gear accuracy for both cases was judged to be grade 8. This vision measurement system can be used for small modulus gears within a certain range.Keywords：vision measurement;system calibration;Pitch deviations;Profile deviations;Gear parameters measurement- 248
- |
- 31
- |
- 4
Published：2021-02-04 -
Formation mechanism of exit chipping in drilling hard-brittle material
Abstract：This study aims to investigate the formation mechanisms of hole exit chipping in diamond drilling of BK7 glass. Based on the microscopic observations of the chipping morphology and fracture mechanics theory， the evolution of extrusion load exerted on the machining surface with crack propagation was explored and its effects on the chipping formation was analyzed. The smooth particle hydrodynamic method （SPH） was applied to numerically simulate the formation process of exit chipping to investigate its formation mechanisms. Results suggest that the exit-chipping morphology could be divided into the incipient crack area and extended crack area. The coupling effects of the extrusion load and bending moment promoted the incipient crack extended in the circumferential direction， which significantly increased the inclination angle of the crack and width of exit chipping， thereby resulting in the occurrence of scattered stripes on the chipping surface. With the SPH method， the simulation of the chipping formation and shedding processes was realized， revealing that the penetration of incipient crack through the entire unpenetrated thickness could reduce the instantaneous cutting force of the tool by 78%. In addition， the cutting force of tool presented periodic fluctuation characteristics with the circumferential progradation of the incipient crack.Keywords：ultra-precision machining;BK7 glass;hole drilling;exit-chipping;smooth particle hydrodynamic method- 172
- |
- 29
- |
- 0
Published：2021-02-04 -
Prediction of milling forces in high-speed milling optical grade SiCp/Al composites
Abstract：Under the conditions of down and up milling， to compare the variation characteristics of milling forces of a polycrystalline-diamond single-tooth end mill with a diamond grain size of 32 μm when milling optical-grade SiCp/Al composites at high speed， the effects of the milling method and milling amounts on milling forces characteristics were studied. The empirical formulae of milling forces were also established based on a multiple linear regression analysis method. An L9（34） orthogonal cutting experiment was conducted to study the influence of milling amounts on the cutting force under the two conditions. A rotary dynamometer was then used to collect the signals of each milling force， and an empirical formula of tangential force Ft， radial force Fr， axial force Fz， and total milling force Ftotal was then established using the multiple linear regression analysis method. Experimental results demonstrated that the milling force calculated using the established regression formula was significantly high， and the average error of the model was within 10%. Under the two milling conditions， the highest level of influence on each milling force was the axial cutting depth， followed by the feed per tooth and cutting speed. To obtain smaller cutting forces， a smaller axial cutting depth should first be selected， followed by a smaller feed per tooth and a higher cutting speed.Keywords：Silicon carbide particle reinforced aluminum matrix composite material;high-speed milling;milling force model;regression analysis;significance analysis- 148
- |
- 25
- |
- 0
Published：2021-02-04 -
Prediction of milling forces in high-speed milling optical grade SiCp/Al composites
Abstract：Under the conditions of down and up milling， to compare the variation characteristics of milling forces of a polycrystalline-diamond single-tooth end mill with a diamond grain size of 32 μm when milling optical-grade SiCp/Al composites at high speed， the effects of the milling method and milling amounts on milling forces characteristics were studied. The empirical formulae of milling forces were also established based on a multiple linear regression analysis method. An L9（34） orthogonal cutting experiment was conducted to study the influence of milling amounts on the cutting force under the two conditions. A rotary dynamometer was then used to collect the signals of each milling force， and an empirical formula of tangential force Ft， radial force Fr， axial force Fz， and total milling force Ftotal was then established using the multiple linear regression analysis method. Experimental results demonstrated that the milling force calculated using the established regression formula was significantly high， and the average error of the model was within 10%. Under the two milling conditions， the highest level of influence on each milling force was the axial cutting depth， followed by the feed per tooth and cutting speed. To obtain smaller cutting forces， a smaller axial cutting depth should first be selected， followed by a smaller feed per tooth and a higher cutting speed.Keywords：Silicon carbide particle reinforced aluminum matrix composite material;high-speed milling;milling force model;regression analysis;significance analysis- 155
- |
- 27
- |
- 0
Published：2021-02-04
Micro/Nano Technology and Fine Mechanics
-
Detection of tiny-moving vehicles in satellite video based on priori-significance
Abstract：Owing to the lack of feature information in satellite videos， obtaining good results using the current feature-based method is difficult. Therefore， in this study， a target detection method based on the image prior information constraint significance is proposed. First， a simple linear iterative clustering algorithm is used to generate a suitable number of superpixels， with compact structures for video frames， and the boundary correlation of each superpixel is calculated based on the human eye significance model. Simultaneously， background modeling is used to generate a motion heat map of the image. Then， motion region estimation is derived based on superpixels. Finally， the two prior constraints are combined to optimize the detection results to obtain a better target detection effect. Different urban road scenes from the “jilin-1” satellite video are taken as examples， and several existing algorithms are compared. Experiments prove that the proposed algorithm can achieve a false alarm rate of less than 10% when the detection accuracy and recall rate reach 85%. The algorithm also has good anti-interference ability in an image environment when the angle deviation of the satellite is considered.Keywords：satellite video;moving vehicle detection;prior information;significance- 177
- |
- 29
- |
- 0
Published：2021-02-04 -
Detection of tiny-moving vehicles in satellite video based on priori-significance
Abstract：Owing to the lack of feature information in satellite videos， obtaining good results using the current feature-based method is difficult. Therefore， in this study， a target detection method based on the image prior information constraint significance is proposed. First， a simple linear iterative clustering algorithm is used to generate a suitable number of superpixels， with compact structures for video frames， and the boundary correlation of each superpixel is calculated based on the human eye significance model. Simultaneously， background modeling is used to generate a motion heat map of the image. Then， motion region estimation is derived based on superpixels. Finally， the two prior constraints are combined to optimize the detection results to obtain a better target detection effect. Different urban road scenes from the “jilin-1” satellite video are taken as examples， and several existing algorithms are compared. Experiments prove that the proposed algorithm can achieve a false alarm rate of less than 10% when the detection accuracy and recall rate reach 85%. The algorithm also has good anti-interference ability in an image environment when the angle deviation of the satellite is considered.Keywords：satellite video;moving vehicle detection;prior information;significance- 131
- |
- 134
- |
- 0
Published：2021-02-04 -
Channel attention and residual concatenation network for image super-resolution
Abstract：Several existing image super-resolution reconstruction methods face challenges owing to information loss between feature channels and during network data transmission. To eliminate this problem， a channel attention and residual concatenation network for image super-resolution is proposed to improve the effect of image super-resolution reconstruction. Initially， shallow feature extraction is performed on the input low-resolution image. Then， the deep features are extracted by the residual concatenation group， and subsequently the attention module is used to adaptively correct the weights of feature channels. The fusion node concatenates， and fuses the shallow and output features of the residual concatenation group to ensure that there is no loss of effective information of the low-resolution image during transmission. Finally， the extracted feature information is reconstructed using a sub-pixel. The experimental results on different benchmark datasets indicates that the proposed method achieves better results in subjective vision and objective index comparison than existing methods. On the Urban100 dataset， the PSNR index of 4 times super-resolution is increased by 0.1 dB. This indicates that the network performs well in image super-resolution reconstruction.Keywords：convolutional neural network;image super-resolution;attention mechanism;concatenation- 319
- |
- 76
- |
- 8
Published：2021-02-04 -
Research on improved VLAD using spatial distribution entropy
Abstract：To improve the accuracy of image retrieval algorithms， an improved VLAD method using spatial distribution entropy is proposed. After apprehending the VLAD features of images， it was discovered that the feature lacks the spatial distribution information of SIFT descriptors. First， the VLAD features in the image were obtained. Second， the spatial distribution information of SIFT descriptors is divided into several sets according to the correspondence between the descriptors and clustering. Third， the spatial distribution entropy is generated in each set. Further， the entropy of all sets is expressed as a spatial distribution entropy vector. Finally， the entropy vector is used to represent the degree of spatial distribution confusion of the descriptor， and the vector is combined with the VLAD. When the codebook size is 64， the experimental results show that the mean average accuracy obtained on the Holidays dataset and the Oxford5k dataset can be increased from 0.521-0.601 and 0.393-0.408， respectively. This method can greatly improve the average accuracy of VLAD features in image retrieval.Keywords：image processing;image retrieval;Vector of Aggragate Locally Descriptor（VLAD）;Scale Invariant Feature Transform（SIFT）;entropy- 162
- |
- 19
- |
- 0
Published：2021-02-04 -
Shortwave infrared visible-light face recognition based on content feature extraction
Abstract：To recognize shortwave-infrared（SWIR） face images according to enrolled visible-light（VIS） face images， a SWIR-VIS face recognition framework based on content feature extraction is proposed. Initially， a SWIR-VIS face image dataset was established. DRIT–an image translation frame–is modified to extract content features more accurately， and consequently obtains better translation results. Then， the content feature extractors in the improved DRIT framework overcome the interference of the modal difference on the recognition. The network used to recognize SWIR faces based on content features was adopted to complete the cross-modal SWIR-VIS face recognition task. The proposed network is evaluated on a self-built SWIR-VIS face image dataset， and compared with the existing widely used methods. Experimental results indicate that the improved DRIT could extract content features more accurately， and consequently the recognition accuracy with content extractors from the improved DRIT model is 12.89% higher than that with the original DRIT content extractors. The recognition accuracy of the proposed framework in the task of SWIR-VIS recognition was 88.86%. The proposed framework can effectively overcome the modality gap and improves the recognition accuracy.Keywords：image translation;short-wave infrared images;face recognition;content feature- 226
- |
- 302
- |
- 2
Published：2021-02-04 -
Fuzzy IAE-UKF integrated navigation method introducing system uncertainty
Abstract：This study addresses the effects of the uncertainty of a combined navigation system of a vehicle-mounted satellite and inertial device on the accuracy of the navigation solution. Based on the statistical characteristics of a filtered innovation estimator and the control characteristics of the ambiguity membership function， a filtering method is designed to suppress the combined navigation error. Considering that prior knowledge of the system is difficult to obtain accurately， the innovation adaptive estimation （IAE） method is first used to adjust the current measurement noise variance online. Then， quantization parameters are determined based on the innovation， and the idea of smoothed weighted residual squared is introduced to provide a weight function that represents system uncertainty. Finally， two improved methods are embedded in the framework of an unscented Kalman filter （UKF） to adjust the noise variance of the system adaptively to form a fuzzy IAE-UKF filter suitable for processing low-cost integrated navigation systems. A vehicle-mounted integrated navigation test is designed and the combined position and speed under different algorithms are compared for accuracy. The results show that the position and speed accuracies of the integrated navigation algorithm in this study were significantly improved. The algorithm realizes the suppression of integrated navigation errors and acts as a filter that can effectively overcome the difficulty of accurately establishing a system model.Keywords：system uncertainty;innovation;fuzzy IAE-UKF filter;integrated navigation;system model- 169
- |
- 31
- |
- 1
Published：2021-02-04 -
Skull identification based on least square canonical correlation analysis
Abstract：To measure the correlation between the skull and the face reliably and improve the skull recognition ability， a skull identification method based on least squares canonical dependency analysis （LSCDA） was proposed. First， a statistical shape model of the skull and facial skin was constructed， and the high-dimensional skull and facial skin were mapped to the low-dimensional shape parameter space. Second， the main relevant information about the skull and facial skin was extracted based on LSCDA， and an overall correlation analysis model was constructed to measure the relationship between the skull and the skin. By considering the difference in the correlations between different regions of the skull， the skull was divided into five regions： the forehead， eyes， nose， mouth， and contour. Based on LSCDA， the main relevant information about the skull and facial skin was extracted and constructed by region. A regional correlation analysis model was used to measure the local detail correlation between the skull and the face. Finally， a global correlation analysis model and a regional correlation analysis model were used to measure the matching relationship between the skull and facial skin， and the matching score between each face in the skull and facial skin database was calculated. The face with the highest matching score yielded the correct recognition result for achieving skull identification. The experimental results reveal that the recognition accuracy of the overall correlation analysis model is 85.2%. The recognition accuracy of the contour region in the single region correlation analysis model is the highest， whereas that of the nose region is the lowest. The correlation analysis established by fusing the five regional models indicates that the recognition accuracy rate is as high as 95.2% and that the method based on regional fusion is better than the overall correlation analysis method.Keywords：image reconstruction;skull recognition;statistical shape model;main related information;least square canonical dependency analysis- 198
- |
- 39
- |
- 2
Published：2021-02-04
Information Sciences
Loading...
0