scholarly journals Strawberry Maturity Recognition Algorithm Combining Dark Channel Enhancement and YOLOv5

Sensors ◽  
2022 ◽  
Vol 22 (2) ◽  
pp. 419
Author(s):  
Youchen Fan ◽  
Shuya Zhang ◽  
Kai Feng ◽  
Kechang Qian ◽  
Yitong Wang ◽  
...  
Keyword(s):  
Recognition Accuracy ◽  
Recognition Rate ◽  
Histogram Equalization ◽  
Recognition Algorithm ◽  
Complex Environments ◽  
Dark Channel ◽  
Testing Accuracy ◽  
Logarithmic Variation ◽  
Enhancement Algorithm ◽  
Better Than

Aiming at the problems of low accuracy of strawberry fruit picking and large rate of mispicking or missed picking, YOLOv5 combined with dark channel enhancement is proposed. In “Fengxiang” strawberry, the criterion of “bad fruit” is added to the conventional three criteria of ripeness, near-ripeness, and immaturity, because some of the bad fruits are close to the color of ripe fruits, but the fruits are small and dry. The training accuracy of the four kinds of strawberries with different ripeness is above 85%, and the testing accuracy is above 90%. Then, to meet the demand of all-day picking and address the problem of low illumination of images collected at night, an enhancement algorithm is proposed to enhance the images, which are recognized. We compare the actual detection results of the five enhancement algorithms, i.e., histogram equalization, Laplace transform, gamma transform, logarithmic variation, and dark channel enhancement processing under the different numbers of fruits, periods, and video tests. The results show that combined with dark channel enhancement, YOLOv5 has the highest recognition rate. Finally, the experimental results demonstrate that YOLOv5 is better than SSD, DSSD, and EfficientDet in terms of recognition accuracy, and the correct rate can reach more than 90%. Meanwhile, the method has good robustness in complex environments such as partial occlusion and multiple fruits.

scholarly journals Unrestricted Face Recognition Algorithm Based on Transfer Learning on Self-Pickup Cabinet

2021 ◽  
Vol 2021 ◽  
pp. 1-12
Author(s):  
Zhixue Liang
Keyword(s):  
Face Recognition ◽  
Transfer Learning ◽  
Recognition Rate ◽  
Recognition Algorithm ◽  
Face Image ◽  
Delivery Device ◽  
Face Images ◽  
Region Extraction ◽  
The Face ◽  
Enhancement Algorithm

In the contactless delivery scenario, the self-pickup cabinet is an important terminal delivery device, and face recognition is one of the efficient ways to achieve contactless access express delivery. In order to effectively recognize face images under unrestricted environments, an unrestricted face recognition algorithm based on transfer learning is proposed in this study. First, the region extraction network of the faster RCNN algorithm is improved to improve the recognition speed of the algorithm. Then, the first transfer learning is applied between the large ImageNet dataset and the face image dataset under restricted conditions. The second transfer learning is applied between face image under restricted conditions and unrestricted face image datasets. Finally, the unrestricted face image is processed by the image enhancement algorithm to increase its similarity with the restricted face image, so that the second transfer learning can be carried out effectively. Experimental results show that the proposed algorithm has better recognition rate and recognition speed on the CASIA-WebFace dataset, FLW dataset, and MegaFace dataset.

scholarly journals Recognition Algorithm of Popular Elements of Ethnic Minority Traditional Clothing Based on PCA

2021 ◽  
Vol 2021 ◽  
pp. 1-10
Author(s):  
Hu Juan
Keyword(s):  
Ethnic Minority ◽  
National Culture ◽  
Euclidean Distance ◽  
Recognition Rate ◽  
Principal Component ◽  
Recognition Algorithm ◽  
Pca Method ◽  
Neural Network Algorithm ◽  
Matching Degree ◽  
Better Than

Image recognition of ethnic minority costumes is helpful for people to understand, carry forward, and inherit national culture. Taking the minority clothing image as the research object, the image enhancement and threshold segmentation are completed; the principal component features of the minority clothing image are extracted by PCA method; and the image matching degree is obtained according to the principle of minimizing the Euclidean distance. Finally, the calculation process of the PCA method is optimized by a wavelet transform algorithm to realize the recognition of popular elements of minority traditional clothing. The comparative experimental results show that the PCA + BP neural network algorithm is better than the other two recognition algorithms in recognition rate and recognition time.

A Novel Face Recognition Algorithm

2013 ◽  
Vol 718-720 ◽  
pp. 2055-2061
Author(s):  
Cai Rang Zhaxi ◽  
Yue Guang Li
Keyword(s):  
Face Recognition ◽  
Recognition Rate ◽  
Feature Space ◽  
Recognition Algorithm ◽  
Analysis Method ◽  
Fisher Linear Discriminant ◽  
Linear Discriminant ◽  
Discriminant Analysis Method ◽  
Pca Method ◽  
Better Than

This paper firstly analyzes the principle of face recognition algorithm, studies feature selection and distance criterion problem, puts forward the defects of PCA face recognition algorithm and LDA face recognition algorithm. According to the deficiencies and shortcomings of PCA face recognition algorithm and LDA face recognition algorithm, this paper proposes a solution -- PCA+LDA. The method uses the PCA method to reduce the dimensionality of feature space, it uses Fisher linear discriminant analysis method to classification, the realization of face recognition. Experiments show that, this method can not only improve the feature extraction speed, but also the recognition rate is better than single PCA method and LDA method.

scholarly journals Detecting Small Chinese Traffic Signs via Improved YOLOv3 Method

2021 ◽  
Vol 2021 ◽  
pp. 1-10
Author(s):  
Baojun Zhang ◽  
Guili Wang ◽  
Huilan Wang ◽  
Chenchen Xu ◽  
Yu Li ◽  
...  
Keyword(s):  
Recognition Accuracy ◽  
Recognition Rate ◽  
Network Computing ◽  
Long Distance ◽  
Effective Technique ◽  
Traffic Signs ◽  
Speed Up ◽  
Small Targets ◽  
Scale Detection ◽  
Better Than

Long-distance detection of traffic signs provides drivers with more reaction time, which is an effective technique to reduce the probability of sudden accidents. It is recognized that the imaging size of far traffic signs is decreasing with distance. Such a fact imposes much challenge on long-distance detection. Aiming to enhance the recognition rate of long-distance small targets, we design a four-scale detection structure based on the three-scale detection structure of YOLOv3 network. In order to reduce the occlusion effects of similar objects, NMS is replaced by soft-NMS. In addition, the datasets are trained and the K-Means method is used to generate the appropriate anchor boxes, so as to speed up the network computing. By using these methods, better experimental results for the recognition of long-distance traffic signs have been obtained. The recognition rate is 43.8 frames per second (FPS), and the recognition accuracy is improved to 98.8%, which is much better than the original YOLOv3.

Intelligent Fuzzy Focusing System For Dark Channel Image Based On Ratio Constraint

2021 ◽  
Author(s):  
Jianhua Lu
Keyword(s):  
Histogram Equalization ◽  
Image Sensor ◽  
Second Step ◽  
Lens System ◽  
Initial Image ◽  
The Third ◽  
Starting Point ◽  
Dark Channel ◽  
And Control ◽  
Better Than

Abstract The starting point of the establishment of the optimized dark channel algorithm is to stabilize the contrast of the two parts of the image (light and dark). The first step of this method is to divide the initial image into two suitable light and dark parts, and then to obtain the contrast data between the two parts, and the second step is to solve the dark area according to the above optimized operation principle. The third step is to exert its accurate management bright features to the extreme to ensure its contrast stability through the operation of double histogram equalization. Modify the unreasonable arrangement of brightness. Take the CCD or CMOS image sensor as an example, the reference will be shown in the above example according to the lens system, and then the intelligent chip of the non-manually operated focusing device will complete the next processing, transmitting the discrimination result to the front system through the motor, and finally focusing. The intelligent fuzzy focusing is composed of the upper and lower computers, the former is the module of collating, collecting materials and processing all the data, the latter is the management and control module of the evaluation results, and the communication part provides communication for the two. Finally, it can be seen that the optimized dark channel algorithm is obviously better than the de-fog algorithm in terms of the effect of information entropy, brightness and average gradient, which makes the detailed characteristics of being obscured by fog more obvious.

The Application of the Fuzzy ART Neural Network Algorithm

2014 ◽  
Vol 513-517 ◽  
pp. 1783-1786 ◽  
Author(s):  
Ming Gu
Keyword(s):  
Neural Network ◽  
Simulation Experiment ◽  
Recognition Rate ◽  
Learning Rule ◽  
Recognition Algorithm ◽  
Fuzzy Art ◽  
Fast Learning ◽  
Average Recognition Rate ◽  
Neural Network Algorithm ◽  
Better Than

An algorithm based on fuzzy ART neural network which can deal with online-learning and recognition of the known and unknown faces at the same time was designed and realized. Based on structure and learning rule of the fuzzy ART system, face recognition algorithm was designed. The simulation experiment results show that average recognition rate of not fast learning is better than fast learning. Not fast learning is accepted to get 89.83% online and 99.42% offline recognition rate.

scholarly journals Radiant Source Signal Recognition Algorithm based on SAE

2018 ◽  
Vol 173 ◽  
pp. 02018
Author(s):  
Ye Wen-qiang ◽  
Yu Zhi-fu ◽  
Zhang Kui ◽  
Wang Hu-bang
Keyword(s):  
Recognition Rate ◽  
Training Model ◽  
Recognition Algorithm ◽  
Identification Algorithm ◽  
Signal Recognition ◽  
Softmax Classifier ◽  
Time Frequency ◽  
Reduction Methods ◽  
Source Signal ◽  
Better Than

Aiming at the shortcomings of traditional radar identification based on artificial judgment and module matching, this paper proposes an intelligent identification algorithm based on joint time-frequency. The radar radiation source signal is transformed by time-frequency, and the processed signal is input into the automatic encoder through different kinds of dimensionality reduction methods, and the pre-training adjustment depth learning model is adopted, and the commonly used softmax classifier is adopted to the pre-training model. Oversee fine school and identification, and finally complete the identification task. The simulation results show that high recognition rate can be achieved by this algorithm, and the joint dimension reduction is better than other methods.

The Recognition and Research of Improved Probabilistic Neural Network

2013 ◽  
Vol 756-759 ◽  
pp. 2819-2824
Author(s):  
Xiao Jing Shang
Keyword(s):  
Neural Network ◽  
Gesture Recognition ◽  
Probabilistic Neural Network ◽  
Recognition Rate ◽  
Recognition Algorithm ◽  
Bp Network ◽  
Training Time ◽  
Neural Network Structure ◽  
Neural Network Algorithm ◽  
Better Than

Probabilistic neural network compared with the traditional BP neural network structure is simpler and it is faster to be identificated, so it is widely used in the field of pattern recognition. This paper is mainly focused on similar gesture recognition research, propose an probabilistic neural network gesture recognition algorithm. The simulation results show that the improved probabilistic neural network algorithm on the recognition rate and training time is better than the traditional BP network.

Skin Detection Model Research Based on Image Enhancement

2011 ◽  
Vol 65 ◽  
pp. 260-263
Author(s):  
Guo Liang Yang ◽  
Zhi Lin Cheng ◽  
Li Zhang
Keyword(s):  
Image Enhancement ◽  
Histogram Equalization ◽  
Original Image ◽  
Skin Model ◽  
Detection Model ◽  
Model Research ◽  
Linear Transform ◽  
Double Skin ◽  
Enhancement Algorithm ◽  
Better Than

Histogram element gradation distribution of the original image concentrates in the low gradation level, and after histogram equalization processed the image is bright and unconspicuous in details. In order to improve the situation, this paper presents an improved image enhancement algorithm .In the algorithm the original image is transformed by the conventional histogram equalization and mapped the histogram equalization processing image as far as possible within the scope of mapping. Then linear transform is used to enhance contrast and apply to mix skin complexion model to extract. Experiments prove that this method is better than double skin model detection at testing results, especially in the eyes and mouth.

Sign in / Sign up

Export Citation Format

Share Document