scholarly journals An Automatic Insect Recognition Algorithm in Complex Background Based on Convolution Neural Network

2020 ◽  
Vol 37 (5) ◽  
pp. 793-798
Author(s):  
Xianrong Zhang ◽  
Gang Chen
Keyword(s):  
Active Contour ◽  
Narrow Band ◽  
Sampling Technique ◽  
Recognition Algorithm ◽  
Fast Method ◽  
Target Region ◽  
Complex Background ◽  
Variable Weight ◽  
Traditional Classification ◽  
Enhancement Algorithm

The existing insect recognition methods mostly segment the target region by traditional classification technology, failing to achieve a high accuracy in complex background. To solve the problem, this paper introduces the morphology-based edgeless active contour strategy to segment insects in complex background. The strategy integrates the morphological operation of gray image, and detects insect contours by narrow-band fast method. To enhance the background diversity of new samples, the authors improved the synthetic minority over-sampling technique (SMOTE) algorithm into a variable weight edge enhancement algorithm. Based on the SMOTE algorithm, the proposed algorithm increases the weight of the edge area as adjacent images are superimposed into a new image, making the background of the new image more complex. Finally, the proposed method was coupled with DenseNet-121 to recognize insects in images with complex background. The results show that the accuracy of the network was nearly 10% higher on the balanced set than on the unbalanced set, suggesting that our method is feasible and accurate.

Iris Localisation Using Fuzzy Centre Detection (FCD) Scheme and Active Contour Snake

2014 ◽  
Vol 69 (6) ◽  
Author(s):  
Masrullizam Mat Ibrahim ◽  
John S. Soraghan ◽  
Nurulfajar Abd Manap
Keyword(s):  
Active Contour ◽  
Iris Recognition ◽  
Image Database ◽  
Target Object ◽  
Recognition Algorithm ◽  
Experimental Result ◽  
Georgia Tech ◽  
Fuzzy Membership Functions ◽  
Main Target ◽  
Snake Algorithm

Iris localisation is a crucial operation in iris recognition algorithm and also in applications, where irises are the main target object. This paper presents a new method to localise iris by using Fuzzy Centre Detection (FCD) scheme and active contour Snake. FCD scheme which consists of four fuzzy membership functions is purposely designed to find a centre of the iris. By using the centre of iris as the reference point, an active contour Snake algorithm is employed to localise the inner and outer of iris boundary. This proposed method is tested and validated with two categories of image database; iris databases and face database.  For iris database, UBIRIS.v1, UBIRIS.v2, CASIA.v1, CASIA.v2, MMU1 and MMU2 are used. Whilst for face databases, MUCT, AT&T, Georgia Tech and ZJUblink databases are chosen to evaluate the capability of proposed method to deal with the small size of the iris in the image database. Based on the experimental result, the proposed method shows promising results for both types of databases, including comparison with the some existing iris localisation algorithm.  

scholarly journals Active Contour Model for Ultrasound Images with Rayleigh Distribution

2014 ◽  
Vol 2014 ◽  
pp. 1-12 ◽  
Author(s):  
Guodong Wang ◽  
Qian Dong ◽  
Zhenkuan Pan ◽  
Ximei Zhao ◽  
Jinbao Yang ◽  
...  
Keyword(s):  
Active Contour ◽  
Active Contour Model ◽  
Speckle Noise ◽  
Rayleigh Distribution ◽  
Ultrasound Images ◽  
Fast Method ◽  
Split Bregman Method ◽  
Split Bregman ◽  
Multiplicative Noises ◽  
Contour Model

Ultrasound images are often corrupted by multiplicative noises with Rayleigh distribution. The noises are strong and often called speckle noise, so segmentation is a hard work with this kind of noises. In this paper, we incorporate multiplicative noise removing model into active contour model for ultrasound images segmentation. To model gray level behavior of ultrasound images, the classic Rayleigh probability distribution is considered. Our model can segment the noisy ultrasound images very well. Finally, a fast method called Split-Bregman method is used for the easy implementation of segmentation. Experiments on a variety of synthetic and real ultrasound images validate the performance of our method.

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.

Algorithm for Automatic Island Segmentation and Identification Based on Narrow Band Deformable Model in UAV Optical Remote Sensing Image

2015 ◽  
Vol 719-720 ◽  
pp. 1209-1216
Author(s):  
En Wei Zhao ◽  
He Meng Yang ◽  
Xiao Jie Wu ◽  
Zeng Zhang
Keyword(s):  
Remote Sensing ◽  
Narrow Band ◽  
Automatic Segmentation ◽  
Recognition Algorithm ◽  
Deformable Model ◽  
Optical Remote Sensing ◽  
Sensing Technology ◽  
Optical Remote Sensing Image ◽  
Initial Segmentation

The island survey is important in economic and strategic field, and in recent years the use of remote sensing technology becomes the mainstream in island investigation. As an effective way for improving the efficiency and accuracy of island survey, the automatic segmentation and recognition algorithm has greater significance. For the difficulty in application of deformed model to high-resolution remote sensing images, the segmentation framework of global initial segmentation and local extractive segmentation based on narrow band deformable model is proposed. Based on the sea and land extraction the island initial segmentation is accomplished, and then the narrow band deformable model is used to increase the accuracy of segmentation. Finally the double rings feature of island is used to improve the quality of the segmentation.

Sign in / Sign up

Export Citation Format

Share Document