Object Recognition with Discriminately Trained Part-Based Model on HOG (Histogram of Oriented Gradients)

Author(s):  
Thanikonda Alekhya ◽  
S. Ranjan Mishra
2018 ◽  
Vol 9 (1) ◽  
Author(s):  
Edgar Utama ◽  
Fikario Yapputra ◽  
Gasim Gasim

<p class="SammaryHeader" align="center"><strong><em>Abstract</em></strong></p><p><em>Many searches have been done for object recognition based on shape using artificial intelligence with various feature extraction. also a lot of research has been done for the introduction of fruit types using various features of extraction and various methods of recognition. This research is identification of mango variant based on shape. Mango is one of the popular fruits and is often consumed by the community. Mango has many variants, each variant of mango usually has a different shape. Mango shape, among others, round with its variants and oval with its variants. the shape of mango is one of the differentiator of each type. Mango used is five variants of mango. The training data comprised 16 images for each variant (a total of 80 images for five mango variants). The test data consisted of 4 images for each variant (a total of 20 images for five mango variants). Train data and test data were obtained by photographing using optical camera, 5 MP secsor resolution, 20 cm photo spacing, and with white background. resize the image used size 50 x 50 pixels on the mango object only. This research uses feature from Histogram of Oriented Gradients (HOG) as training input and testing of recognition method. Recognition method using Artificial Neural Network with back propagation algorithm. Accuracy of recognition that can be achieved in this study is 90%.</em></p><p><strong><em>Keywords </em></strong><em>:<strong> </strong>Object recognition, Shape, HOG, Artificial Neural Network</em></p><p align="center"><strong> </strong></p><p align="center"><strong><em>Abstrak</em></strong><em></em></p><p><em>Banyak penelian yang pernah dilakukan untuk objek recognition berdasarkan shape menggunakan kecerdasan buatan dengan berbagai ekstraksi ciri. juga banyak penelitian yang pernah dilakukan untuk pengenalan jenis buah menggunakan berbagai fitur ekstraksi dan berbagai metode recognition. Penelitian ini adalah identifikasi varian mangga berdasarkan shape. Mangga merupakan salah satu buah yang popular dan sering dikonsumsi oleh masyarakat. Mangga memiliki banyak varian,  tiap varian dari mangga biasanya memiliki bentuk yang berbeda. Bentuk mangga antara lain bulat dengan variannya dan lonjong dengan variannya. the shape mangga merupakan salah satu pembeda dari masing-masing jenis. Mangga yang digunakan adalah lima varian mangga. Data latih terdiri 16 citra untuk tiap variannya (total 80 citra untuk lima varian mangga). Data uji terdiri dari 4 citra untuk tiap variannya (total 20 citra untuk lima varian mangga). Data latih dan data uji didapatkan dengan cara difoto menggunakan kamera optik, resolusi secsor 5 MP, jarak foto 20 cm, dan dengan latar putih. resize citra yang digunakan berukuran 50 x 50 piksel pada bagian objek mangga saja. Research ini menggunakan feature from Histogram of Oriented Gradients (HOG) sebagai input pelatihan dan pengujian metode recognition. Metode recognition menggunakan Jaringan Saraf Tiruan dengan algoritma propagasi balik. Akurasi pengenalan yang dapat dicapai dalam penelitian ini adalah sebesar 90%.</em></p><strong><em>Kata kunci </em></strong><em>:<strong> </strong>Pengenalan<strong> </strong>Objek, bentuk , HOG, Jaringan Saraf Tiruan</em>


GeroPsych ◽  
2010 ◽  
Vol 23 (3) ◽  
pp. 169-175 ◽  
Author(s):  
Adrian Schwaninger ◽  
Diana Hardmeier ◽  
Judith Riegelnig ◽  
Mike Martin

In recent years, research on cognitive aging increasingly has focused on the cognitive development across middle adulthood. However, little is still known about the long-term effects of intensive job-specific training of fluid intellectual abilities. In this study we examined the effects of age- and job-specific practice of cognitive abilities on detection performance in airport security x-ray screening. In Experiment 1 (N = 308; 24–65 years), we examined performance in the X-ray Object Recognition Test (ORT), a speeded visual object recognition task in which participants have to find dangerous items in x-ray images of passenger bags; and in Experiment 2 (N = 155; 20–61 years) in an on-the-job object recognition test frequently used in baggage screening. Results from both experiments show high performance in older adults and significant negative age correlations that cannot be overcome by more years of job-specific experience. We discuss the implications of our findings for theories of lifespan cognitive development and training concepts.


Sign in / Sign up

Export Citation Format

Share Document