A Novel Method for Handicrafts Design Based on Fusion of Multi-Intelligent Decision Algorithm

With the rapid development of artificial intelligence, handicraft design has developed from artificial design to artificial intelligence design. Traditional handicraft design has the problems of long time consumption and low output, so it is necessary to improve the process technology. Artificial intelligence technology can provide optimized design steps in handicraft design and improve design efficiency and process level. Handicrafts are regarded as important social products and exist in people’s daily life. In the current society, many people do handicrafts and there are major exhibitions. Furthermore, the display of handicrafts is also very grand and shocking. In the design of handicrafts, the traditional design method cannot completely keep up with the production speed and efficiency of handicrafts. Therefore, this paper adopts the fusion multi-intelligent decision algorithm of multi-node branch design in the design method of handicraft. The algorithm model combination is used to analyze and design the layout of the handicraft, which speeds up the design efficiency and production of the handicraft. In this paper, two intelligent algorithms will be used for fusion; they are genetic algorithm and GA-PSO fusion algorithm obtained by particle swarm optimization and they are embedded in handicraft design method for application through mathematical model construction and function construction. After comparing the performance parameter index data of three intelligent algorithms and GA-PSO fusion algorithm, it is obtained that GA-PSO fusion algorithm is 97% correct and has 82% readability, 72% robustness, and 61% structure, making it have better important indicators. Four algorithms optimize each design problem in all aspects of handicraft design at present. Design efficiency, image distribution rate, image optimization degree, and image clarity are compared by simulation experiments. Compared with three intelligent algorithms, traditional design methods, and manual design methods, GA-PSO fusion algorithm can effectively improve the design method and design effect of handicrafts with 92.1% design efficiency, 82.7% image distribution rate, 94.3% image optimization degree, and 84% layout void rate. Finally, the space complexity experiment of four algorithms shows that GA-PSO algorithm can achieve 9.73 dispersion with 11.42 space complexities, which makes the dimension reduction relatively stable, and the algorithm can maintain stability in the design and application of handicrafts.

Optimisasi Dosis dan Kualitas Citra CT-Scan untuk Variasi Pitch sebagai Upaya Proteksi Radiasi di Rumah Sakit Umum Pusat Adam Malik Medan

Optimization is an effort to ensure patient radiation safety and is the main action in overcoming concerns about CT-Scan radiation exposure. This led to the emergence of various measures to reduce the dose. This study aims to obtain a minimal dose with a high-quality image. Optimization efforts were carried out by the radiology team at Adam Malik Hospital Medan using a 16 slice GE CT-Scan and a water phantom with a diameter of 16 and 32 cm and an image quality questionnaire form. Collected data by observing the head, chest, and abdomen CT-Scan in adult patients (≥15 years). The data taken is the value of CTDI vol and DLP for a year. Then a water phantom scan was carried out with the head protocol using pitch parameters 0.562 and 0.938. The chest and abdomen use pitches of 1.375 and 1.75. The results obtained were evaluated and applied to patients, then filled in the image quality questionnaire scores. The results of CTDI_vol and DLP values with 16 and 32 cm water phantom scans showed a decrease in the dose value; for pitch 0.938, it was 1.6% lower than pitch 0.562, and pitch 1.75 was 1.2% lower, compared to pitch 1.375. For CT head examination using a pitch of 0.963, the CTDI_vol value was 1.5%, and DLP was 2%. For chest using a pitch of 1.75, CTDI_vol values were 1.3% and DLP 2%, while abdominal examination with a pitch of 1.75 obtained CTDI_vol values 1.8% and DLP 1.4%. From these three results, the CTDI_vol and DLP values were higher than the national DRL values. The value obtained is higher than the national DRL due to differences in the phantom test protocol with clinical implementation and the lack of accuracy in using other parameters. Changes in scan parameters are not comprehensive. Obtained a score of 3 in the questionnaire form stating that the radiology doctor can still interpret the image. This study concluded that it could make optimization efforts by changing the pitch parameter by paying attention to other parameters without reducing the quality of the image interpreted by the radiologist.

Training Image Optimization Method Based on Convolutional Neural Network and Its Application in Discrete Fracture Network Model Selection

Training image (TI) is important for multipoint statistics simulation method (MPS), since it captures the spatial geological pattern of target reservoir to be modeled. Generally, one optimal TI is selected before applying MPS by evaluating the similarities between many TIs and the well interpretations of target reservoir. In this paper, we propose a new training image optimization approach based on the convolutional neural network (CNN). First, candidate TIs were randomly sampled several times to obtain the sample dataset. Then, the CNN was used to conduct transfer learning for all samples, and finally, the optimal TI of the conditioning well data is selected through the trained CNN model. By taking advantage of the strong learning ability of CNN in image feature recognition, the proposed method can automatically identify differences in spatial features between the conditioning well data and the samples of the training image. Hence, it effectively resolves the difficulty of spatial matching between discrete datapoints and grid structures. We demonstrated the applicability of our model via 2D and 3D training image selection examples. The proposed methods effectively selected the appropriate TI, and then the pretreatment techniques for improving the accuracy of continuous TI selection were achieved. Moreover, the proposed method was successfully applied to training image selection of a discrete fracture network model. Finally, sensitivity analysis was carried out to show that sufficient conditioning data volume can reduce the uncertainty of the optimization results. By comparing with the improved MDevD method, the advantages of the new method are verified in terms of efficiency and reliability.