1. bookAHEAD OF PRINT
Informacje o czasopiśmie
License
Format
Czasopismo
eISSN
2444-8656
Pierwsze wydanie
01 Jan 2016
Częstotliwość wydawania
2 razy w roku
Języki
Angielski
Otwarty dostęp

Multimedia sensor image detection based on constrained underdetermined equation

Data publikacji: 15 Jul 2022
Tom & Zeszyt: AHEAD OF PRINT
Zakres stron: -
Otrzymano: 10 Apr 2022
Przyjęty: 05 Jun 2022
Informacje o czasopiśmie
License
Format
Czasopismo
eISSN
2444-8656
Pierwsze wydanie
01 Jan 2016
Częstotliwość wydawania
2 razy w roku
Języki
Angielski
Introduction

Thanks to the technical integration of radio frequency (RF) communication technology, digital electronic technology and micro electro mechanical systems (MEMS), micro, cheap, low-power and multifunctional wireless sensor nodes can be realized, and wireless sensor networks (WSN s) have been developed rapidly and widely used [1]. Sensor nodes in wireless sensor networks have the ability of environment awareness, data processing and communication. Wireless networks are formed between nodes through multi hop, and complete large-scale and complex monitoring tasks in a cooperative way. Wireless sensor networks connect people with rich and real information in the real world more directly and closely, and provide sufficient development space and effective solutions for the increasing application scenarios and requirements in the current era of Internet of things [2]. At present, wireless sensor networks have been widely used in national defense support, environmental monitoring, health care, smart home and smart city, and have played a positive role in social construction, nature protection and personal development. In order to realize the perception and acquisition of multimedia information, sensor nodes in WMSN s are generally equipped with CMOS cameras and micro microphones in addition to scalar sensors with simple environmental data acquisition function [3]. WMSN s integrates technologies in many fields such as digital signal processing, communication, network, control and statistics. It can realize the retrieval of video stream, audio stream and still image and the real-time transmission of information, and store, process, correlate and fuse the information from different services. Due to the introduction of multimedia processing technology, WMSN s can realize higher precision monitoring, so that end users can obtain more objective and comprehensive information, and then complete complex advanced monitoring tasks [4]. At present, there are two common methods: machine vision detection and manual detection. Among them, the former is greatly affected by experience and subjective factors, so machine vision detection has gradually become the mainstream. However, because there are dense parallel grid lines on the surface of solar cells, and the defects are subtle and hidden, the algorithm has become a key factor in the field of machine vision to identify the surface defects of solar cells [5]. In addition, the image acquisition of battery is greatly affected by light, which makes it difficult to identify defects. In order to remove these effects, the commonly used detection methods in time domain, frequency domain and wavelet domain need to introduce preprocessing technologies, such as LBP algorithm, Hough transform, normalization and frequency domain filtering. However, these technologies set the empirical threshold on the data after image transformation to remove the influence of illumination and parallel grid, which inevitably increases the complexity of the algorithm, and it is easy to eliminate slight scratches as the background and increase the false detection rate. In addition, considering the differences of parallel grids of battery chips from different manufacturers, the empirical threshold needs to be estimated accordingly, which brings inconvenience to use [6]. Aiming at this research problem, Huang B and others fused the color, direction and gray features of the image, which is the most typical method for significance detection of fused low-level features [7]. Ambareesh S and others first used the face detection algorithm to construct the face features reflecting the image face region, and then combined them with low-level features to improve the accuracy of significance detection. The above methods all use the linear fusion method with equal weight for feature fusion, and the fusion parameters of the model are constructed manually [8]. Kim B S and others used the model to quote the features used in other significance detection related work, calculated the low-level, medium-level and high-level features of 33 images, and then combined with the information in the eye movement database to construct training samples, and regarded the process of fusing features and predicting significant areas as a binary classification problem, a nonlinear feature fusion model is trained by SVM method [9]. Xiao D and others characterized the detection results of some other significance detection models [10]. Sharma R P and others' models are used to learn the model parameters with the methods of expression, SVM and boosting. The detection results are better than most other models. This kind of feature fusion model constructed with the idea of statistical modeling can automatically learn feature fusion parameters and avoid too much human intervention in the process of model construction [11].

Based on the current research, aiming at the problem of low reliability of image sensor quality detection results, this paper establishes an image recognition model based on Laplacian operator image recognition technology and MATLAB software to realize the functions of image acquisition, image edge intelligent recognition, image feature information intelligent extraction and analysis. The number of samples in this experiment is 50. The products are numbered and grouped. Since the base of the product warpagetest needs to be removed from the surface of the product image sensor after the base installation process, the whole product will be scrapped and cannot flow to the next process for production. Referring to the existing production materials, it is found that the existing production processes that may have a great impact on the flatness of the image sensor are false hardening, heating hardening, base installation, calibration and bonding, soldering and testing, so each group shall set the measurement. Among them, No. 22, No. 39 and No. 40 products are poor due to changes in physical properties in subsequent production, and cannot flow to the next process for production. Therefore, the data of No. 22, No. 39 and No. 40 are missing in the test data. A total of 133 groups of experimental data were obtained in this experimental test. The results showed that the standard deviations of group 1, group 2, group 3 and group 4 were 0.172, 0.125 and 0.304 respectively. That is, after the same product is treated by false hardening, heating hardening and base installation processes, the false hardening and heating hardening processes have relatively little impact on the surface flatness of the image sensor. After the base opening and installation process, the surface flatness of the image sensor begins to change. The standard deviations of the measurement results of group 1 and groups 4, 5, 6 and 7 were 0.304, 0.381, 0.391 and 0.514 respectively. That is, after the base installation, calibration, bonding, soldering and testing of the product, the surface flatness of the image sensor and the initial surface flatness have changed greatly. Therefore, production technicians can focus on the process that starts to change, that is, the base installation process. The improved method has the advantages of automatic operation, low error, low error rate and so on.

Method
Current situation of image quality detection

The time interval is set as Δt; The space step size is 1.(I, j) represents the discrete points in the image, and the iterative formula after the dispersion of the model in this paper is as follows:

The camera quality is the most important part of the camera quality of the mobile phone. The case studied in this paper is model h mobile phone camera, and the first pass rate of good products is only 80%. In order to meet the 95% pass rate of good products required by customers, it is necessary to comprehensively improve the processes that have a great impact on product quality, so as to improve the first pass rate of products, achieve quality objectives and reduce manufacturing failure costs [14].

According to the requirements of the existing quality control system, when the first batch of each equipment is produced in each shift, process and process, after a certain number of products are trial produced, stop the equipment and randomly select three products for quality inspection. If all three random products are qualified, the equipment in this shift and process can be mass produced; If there are unqualified products among the three products, the equipment shall be debugged, and after a certain number of products are produced again, three products shall be randomly selected for quality inspection. Until three products are randomly selected as qualified products.

The main quality inspection item studied in this paper is WarpageTest, which measures the surface flatness of image sensor. In the production process of mobile phone camera, if the surface flatness of image sensor exceeds 5μm, its imaging photosensitive performance will be affected. Therefore, the production process should ensure that the height difference of surface flatness does not exceed 5μm. At present, with the help of VR30003D profilometer, the surface image of the image sensor is obtained, and the surface height difference of the image sensor is calculated with the help of 3D profilometer supporting analysis software. The specific analysis steps are as follows.

Select the photosensitive area of the image sensor as the reference plane, automatically calculate the average height of the reference plane as the 0 degree reference line of the whole area, and set the display height color information.

In the selected area, draw two straight lines from top left to bottom right and from bottom left to top right respectively, with a line width of 30, that is, draw 30 straight lines with an average width at the same time to obtain the height line of the area through which the 30 straight lines pass, and calculate the average value as the height line of the straight lines drawn from top left to bottom right and from bottom left to top right.

Manually select the midpoint of the drawn line as the reference point, and manually select the starting point, 1 / 4 point, 3 / 4 point and end point of the drawn line in turn. Finally, the software can automatically obtain the height difference between the reference point and the last 4 points. Repeat for another line.

The operator shall check the final height difference data, and if the maximum value does not exceed 5μm, it will be judged as good product.

Existing problems in image quality detection

Through communication with on-site operators and quality supervisors, we use brainstorming and other methods to summarize many shortcomings of existing measurement methods, and sort out the existing problems based on man-machine material method, environmental measurement and fishbone diagram analysis method.

The existing image measurement methods are not scientific enough and human subjectivity is strong. Image recognition is mainly based on photosensitive areas, and there is a lack of scientific extraction methods. The existing detection adopts manual marking and setting center points to obtain the detection interval, which can not fully and accurately reflect the overall plane image characteristics, and it is easy to miss detection [15].

The detection link is not scientific enough. Due to the limitation of the existing production process, the base installation process cannot be tested directly on the premise of keeping the surface flatness of the image sensor unchanged. The War page Test only after the patch process can not ensure the reliability of the detection quality.

Intelligent extraction of feature information of collected images

The 3D profilometer can obtain the maximum height heightmax and minimum height heightmin in the measurement area, in which red indicates the area with higher height min and dark blue indicates the area with lower height. For the height feature information extraction of the whole plane, we only need to find out the maximum color value (red value) and the minimum color value (blue value) in the HSL color model matrix [h, s, l] after image digitization. The value difference between the two is the height feature information gap in the whole plane. Therefore, the height matrix component value of the whole measurement area is: height=H×heightmaxheightminHmaxHmin height = H \times {{heigh{t_{\max }} - heigh{t_{\min }}} \over {{H_{\max }} - {H_{\min }}}}

After extracting the height feature information of the measurement area of the image sensor and obtaining the plane height matrix, the original two-dimensional plane image can be transformed into a three-dimensional image, and the data of the plane height feature information can be analyzed by using the statistical principle.

Characteristic information and data analysis of collected images

After extracting the height information of the main measurement area of the collected image, the height information is analyzed, and the flatness evaluation of the measurement area is obtained by using the standard deviation tool in the principle of data statistics. The calculation formula is: SD=1Ni=1N(heightheight¯)2 SD = \sqrt {{1 \over N}\sum\limits_{i = 1}^N {{{\left( {height - \overline {height} } \right)}^2}} } Among them, N=nm;height¯=i=1NheightN N = nm;\overline {height} = {{{\sum\limits_{i = 1}^N {height} } \over N}_ \circ } N, m is the number of vertical and horizontal pixels in the main imaging area of the image; height¯ \overline {{\rm{height}}} is the average value of the height information of the main imaging area of the image.

Results and analysis
Setting of product measurement experiment

To find the threshold of height information measured by the improved method, judge whether the product is good or not, and identify the process that significantly affects the surface smoothness of the image sensor. Custom. The number of samples for this test is 50. Products are numbered and grouped. After the installation process, the base needs to be removed from the surface of the product image sensor, so that the product is completely scrapped and cannot flow into the next production process. Existing production materials include existing production processes such as forging hardening, thermal hardening, base mounting, adjusting connections, welding and testing, which can significantly affect the smoothness of the image sensor. Therefore, the measurement Settings for each group are shown in Table 1. The first four groups of tests are war Page tests of single product after different processes, and the last three groups of tests are WAR Page tests of different products. Through various processes. At the beginning of production, due to the uneven surface of the image sensor, 9 reserved products were used as a reference.

Experimental settings

Group number Number Variable setting Naming method
1 1–51 Conduct War pageTest after the pressure welding 1–51.1
2 1–18 Conduct War pageTest after the false hardening process 1–17.23
3 1–18 After the heating and hardening process, 1–17.29
4 1–11 Conduct War pageTest after the base 1–10.39
5 18–27 Conduct War pageTest after the calibration 18–28.49
6 29–37 Conduct War pageTest after soldering 29–38.59
7 39–51 Conduct War pageTest after the product test procedure 39–50.69
8 51–58 It is found that the flatness of the image sensor is poor 51–59.79
Analysis of measured experimental data of products

The improved war page test method was used to measure the products of each experimental group. Among them, No. 22, No. 39 and No. 40 products are poor due to changes in physical properties in subsequent production, and cannot flow to the next process for production. Therefore, the data of No. 22, No. 39 and No. 40 are missing in the test data. A total of 133 groups of experimental data were obtained in this experimental test.

Determination of process affecting surface flatness of image sensor

Based on the patch process in production, the war page test was inspected as good, but after the final assembly of the product, the test link found that the product had a poor quality accident of the whole product due to the change of the surface flatness of the image sensor. The defective product is No. 51–59 in the experiment.

Therefore, in order to completely avoid the recurrence of such quality accidents, production technicians need to determine the processes that have a great impact on the surface flatness of the image sensor in each process after the die bonding process, so as to narrow the spot inspection range and find out the root cause of the impact on the surface flatness of the image sensor.

After improving the measurement method of War page Test, we can extract the height information and analyze the data of the whole plane of the image sensor, that is, we can find out the processes that have a great impact on the surface flatness of the image sensor. Compare the measurement results of War page Test between group 1 and other groups before and after assembly in each process, and compare the measurement results. The difference significance test is carried out, and the results are shown in Table 2.

Paired t-test of measured results

Pair Mean value Standard deviation T value p
Groups 1 and 2 0.108 0.172 2.610 0.030
Groups 1 and 3 0.047 0.125 1.595 0.081
Groups 1 and 4 −0.176 0.304 −1.920 0.002
Groups 1 and 5 −0.196 0.381 −1.631 0.036
Groups 1 and 6 −0.125 0.391 −1.025 0.018
Groups 1 and 7 −0.263 0.514 −1.621 0.041

The significant difference between the paired groups was reflected by the paired t-test of each group. The test results showed that there was a very significant difference between group 1 and group 4 (P = 0.003 < 0.01), and there was a significant difference between group 1 and group 2, group 5, group 6 and group 7 (P = 0.030 < 0.05, P = 0.036 < 0.05, P = 0.018 < 0.05, P = 0.041 < 0.05), but there was no significant difference between group 1 and group 3 (P = 0.081 > 0.05). The following experimental conclusions can be drawn by comparing the measurement results of the last 6 groups with the measurement results of the first group.

The standard deviations of group 1, group 2, group 3 and group 4 were 0.172, 0.125 and 0.304 respectively. Taking the measurement results of group 1, group 2 and group 4 as an example, the comparison diagram is shown in Figure 1 and Figure 2. That is, after the same product is treated by false hardening, heating hardening and base installation processes, the false hardening and heating hardening processes have relatively little impact on the surface flatness of the image sensor. After the base opening and installation process, the surface flatness of the image sensor begins to change.

Figure 1

Comparison of measurement results of group 1 and group 2

Figure 2

Comparison of measurement results between group 1 and group 4

Compared with the measurement results of group 1 and group 2, the data of group 2 is generally less than that of group 1, that is, the surface flatness of the image sensor gradually becomes stable after the false hardening process of the same product, which shows that the false hardening process helps the surface of the image sensor become flat. Since the false hardening process is a short-term low-temperature heating of the product, which has no impact on other properties of the product, in order to flatten the surface of the image sensor, the false hardening process can be added after individual processes that have a great impact on the surface flatness of the image sensor.

The standard deviations of the measurement results of group 1 and groups 4, 5, 6 and 7 were 0.304, 0.381, 0.391 and 0.514 respectively. That is, after the base installation, calibration, bonding, soldering and testing of the product, the surface flatness of the image sensor and the initial surface flatness have changed greatly. Therefore, production technicians can focus on the process that starts to change, that is, the base installation process.

Effect analysis of improved method and original measurement method

The improved method has the advantages of automatic operation, low error, low error rate and so on. Compared with the original measurement method, the advantages after improvement are summarized as follows.

Without the participation of operators, the program algorithm can automatically carry out image recognition, analysis and processing, output results and other processes, completely avoiding the quality control problems caused by the operator's operation errors and the interference of human subjective factors in the detection process.

Compared with the original measurement method, the improved method can output the height feature information of the whole plane, rather than just manually drawing the straight line area, and can more comprehensively reflect the surface flatness feature of the whole image sensor.

The time of the whole test process is greatly shortened. The whole analysis process can realize the self operation of the computer. The time of the whole war page test mainly depends on the image acquisition time of the 3D profilometer, and the test time is shortened from 5.1min to 2min.

After improvement, the products after each process shall be subject to war page test, and the surface flatness shall be compared and analyzed longitudinally through multiple groups of data to make up for the deficiency of war page test measurement method before improvement.

Conclusion

In this paper, multimedia sensor image detection based on constrained underdetermined equation is proposed. Aiming at the problem of low reliability of image sensor quality detection results, based on Laplacian operator image recognition technology and Matlab software, an image recognition model is established to realize the functions of image acquisition, image edge intelligent recognition, image feature information intelligent extraction and analysis. The number of samples in this experiment is 50. The products are numbered and grouped. Because the base on the surface of the product image sensor needs to be removed after the base installation process, the whole product will be scrapped and cannot flow to the next process for production. Referring to the existing production materials, it is found that the existing production processes that may have a great impact on the flatness of the image sensor are false hardening, heating hardening, base installation, calibration and bonding, soldering and testing, so each group shall set the measurement. After improving the measurement method of War page Test, we can extract the height information and analyze the data of the whole plane of the image sensor, that is, we can find out the processes that have a great impact on the surface flatness of the image sensor. Compare the measurement results of War page Test between group 1 and other groups before and after assembly in each process, and compare the measurement results. The significance of difference was tested. Through experimental comparison, it is proved that the improved detection method can more accurately reflect the quality characteristics of products. However, the image recognition algorithm proposed in this paper is currently an independent detection program, and the operator still needs to operate the 3D contour measuring instrument to collect the product image. The next improvement direction is to connect the detection program with the image acquisition program to realize the intellectualization of the whole detection link.

Figure 1

Comparison of measurement results of group 1 and group 2
Comparison of measurement results of group 1 and group 2

Figure 2

Comparison of measurement results between group 1 and group 4
Comparison of measurement results between group 1 and group 4

Paired t-test of measured results

Pair Mean value Standard deviation T value p
Groups 1 and 2 0.108 0.172 2.610 0.030
Groups 1 and 3 0.047 0.125 1.595 0.081
Groups 1 and 4 −0.176 0.304 −1.920 0.002
Groups 1 and 5 −0.196 0.381 −1.631 0.036
Groups 1 and 6 −0.125 0.391 −1.025 0.018
Groups 1 and 7 −0.263 0.514 −1.621 0.041

Experimental settings

Group number Number Variable setting Naming method
1 1–51 Conduct War pageTest after the pressure welding 1–51.1
2 1–18 Conduct War pageTest after the false hardening process 1–17.23
3 1–18 After the heating and hardening process, 1–17.29
4 1–11 Conduct War pageTest after the base 1–10.39
5 18–27 Conduct War pageTest after the calibration 18–28.49
6 29–37 Conduct War pageTest after soldering 29–38.59
7 39–51 Conduct War pageTest after the product test procedure 39–50.69
8 51–58 It is found that the flatness of the image sensor is poor 51–59.79

Yang J, Jia Z, X Lü, et al. Digital image biological detection technology based on the porous silicon periodic crystals film[J]. Optoelectronics Letters, 2021, 17(9):552–557. YangJ JiaZ X Digital image biological detection technology based on the porous silicon periodic crystals film [J] Optoelectronics Letters 2021 17 9 552 557 10.1007/s11801-021-0176-5 Search in Google Scholar

Ma Z, Guo D. Discrete-Time Recurrent Neural Network for Solving Bound-Constrained Time-Varying Underdetermined Linear System[J]. IEEE Transactions on Industrial Informatics, 2020, PP(99):1–1. MaZ GuoD Discrete-Time Recurrent Neural Network for Solving Bound-Constrained Time-Varying Underdetermined Linear System [J] IEEE Transactions on Industrial Informatics 2020 PP 99 1 1 10.1109/TII.2020.3012993 Search in Google Scholar

Smith C B, Hernandez E M. Non-negative and sparsity constrained inverse problems in damage identification - Application to a full-scale 3D truss[J]. Mechanical systems and signal processing, 2020, 140(Jun.):106648.1–106648.17. SmithC B HernandezE M Non-negative and sparsity constrained inverse problems in damage identification - Application to a full-scale 3D truss [J]. Mechanical systems and signal processing 2020 140 Jun. 106648.1 106648.17 10.1016/j.ymssp.2020.106648 Search in Google Scholar

Sun J, Zhao J, Wang K. Online Surge Detection Method Based on Axial Displacement Sensor of MSCC[J]. IEEE Sensors Journal, 2019, PP(99):1–1. SunJ ZhaoJ WangK Online Surge Detection Method Based on Axial Displacement Sensor of MSCC [J] IEEE Sensors Journal 2019 PP 99 1 1 10.1109/JSEN.2019.2909311 Search in Google Scholar

Kumar A R, Sivagami A. Fuzzy based malicious node detection and security-aware multipath routing for wireless multimedia sensor network[J]. Multimedia Tools and Applications, 2020, 79(19):14031–14051 KumarA R SivagamiA Fuzzy based malicious node detection and security-aware multipath routing for wireless multimedia sensor network [J] Multimedia Tools and Applications 2020 79 19 14031 14051 10.1007/s11042-020-08631-0 Search in Google Scholar

Han C, Zhang S, Zhang B, et al. A Distributed Image Compression Scheme for Energy Harvesting Wireless Multimedia Sensor Networks[J]. Sensors, 2020, 20(3):667. HanC ZhangS ZhangB A Distributed Image Compression Scheme for Energy Harvesting Wireless Multimedia Sensor Networks [J] Sensors 2020 20 3 667 10.1109/MSN48538.2019.00075 Search in Google Scholar

Huang B, Xu T, Shen Z, et al. BSCF: Learning background suppressed correlation filter tracker for wireless multimedia sensor networks[J]. Ad Hoc Networks, 2021, 111(2):102340. HuangB XuT ShenZ BSCF: Learning background suppressed correlation filter tracker for wireless multimedia sensor networks [J] Ad Hoc Networks 2021 111 2 102340 10.1016/j.adhoc.2020.102340 Search in Google Scholar

Ambareesh S, Madheswari A N. HRDSS-WMSN: A Multi-objective Function for Optimal Routing Protocol in Wireless Multimedia Sensor Networks using Hybrid Red Deer Salp Swarm algorithm[J]. Wireless Personal Communications, 2021, 119(1):117–146. AmbareeshS MadheswariA N HRDSS-WMSN: A Multi-objective Function for Optimal Routing Protocol in Wireless Multimedia Sensor Networks using Hybrid Red Deer Salp Swarm algorithm [J] Wireless Personal Communications 2021 119 1 117 146 10.1007/s11277-021-08201-z Search in Google Scholar

Kim B S, Kim S, Kim K H, et al. Adaptive Real-Time Routing Protocol for (m, k)-Firm in Industrial Wireless Multimedia Sensor Networks[J]. Sensors, 2020, 20(6):1633. KimB S KimS KimK H Adaptive Real-Time Routing Protocol for (m, k)-Firm in Industrial Wireless Multimedia Sensor Networks [J] Sensors 2020 20 6 1633 10.3390/s20061633714660232183403 Search in Google Scholar

Xiao D, Li M, Wang M, et al. Low-cost and high-efficiency privacy-protection scheme for distributed compressive video sensing in wireless multimedia sensor networks[J]. Journal of Network and Computer Applications, 2020, 161(4):102654. XiaoD LiM WangM Low-cost and high-efficiency privacy-protection scheme for distributed compressive video sensing in wireless multimedia sensor networks [J] Journal of Network and Computer Applications 2020 161 4 102654 10.1016/j.jnca.2020.102654 Search in Google Scholar

Sharma R P, Dey S. A comparative study of handcrafted local texture descriptors for fingerprint liveness detection under real world scenarios[J]. Multimedia Tools and Applications, 2021, 80(7):9993–10012. SharmaR P DeyS A comparative study of handcrafted local texture descriptors for fingerprint liveness detection under real world scenarios [J] Multimedia Tools and Applications 2021 80 7 9993 10012 10.1007/s11042-020-10136-9 Search in Google Scholar

Mukherjee A, Goswami P, Yang L. DAI based wireless sensor network for multimedia applications[J]. Multimedia Tools and Applications, 2020(5):1–15. MukherjeeA GoswamiP YangL DAI based wireless sensor network for multimedia applications [J] Multimedia Tools and Applications 2020 5 1 15 10.1007/s11042-020-08909-3 Search in Google Scholar

Aslam A, Curry E. A Survey on Object Detection for the Internet of Multimedia Things (IoMT) using Deep Learning and Event-based Middleware: Approaches, Challenges, and Future Directions[J]. Image and Vision Computing, 2020, 106(10):104095. AslamA CurryE A Survey on Object Detection for the Internet of Multimedia Things (IoMT) using Deep Learning and Event-based Middleware: Approaches, Challenges, and Future Directions [J] Image and Vision Computing 2020 106 10 104095 10.1016/j.imavis.2020.104095 Search in Google Scholar

Derya Arslan. The Comparison Study of Hybrid Method with RDTM for Solving Rosenau-Hyman Equation[J]. Applied Mathematics and Nonlinear Sciences, 2020, 5(1) : 267–274. ArslanDerya The Comparison Study of Hybrid Method with RDTM for Solving Rosenau-Hyman Equation [J] Applied Mathematics and Nonlinear Sciences 2020 5 1 267 274 10.2478/amns.2020.1.00024 Search in Google Scholar

Volkan Ala and Khanlar R. Mamedov. On Basis Property of Root Functions For a Class Second Order Differential Operator[J]. Applied Mathematics and Nonlinear Sciences, 2020, 5(1) : 361–368. AlaVolkan MamedovKhanlar R. On Basis Property of Root Functions For a Class Second Order Differential Operator [J] Applied Mathematics and Nonlinear Sciences 2020 5 1 361 368 10.2478/amns.2020.1.00034 Search in Google Scholar

Polecane artykuły z Trend MD

Zaplanuj zdalną konferencję ze Sciendo