Wyniki wyszukiwania - BazTech

1

Novel Feature Extraction for Pineapple Ripeness Classification

Wang Hui Hui, Chai Sze Ye

Journal of Telecommunications and Information Technology

|

2022

|

nr 1

14--22

EN

A novel feature extraction method has been proposed to improve the accuracy of the pineapple ripeness classification process. The methodology consists of six stages, namely: image acquisition, image pre-processing, color extraction, feature selection, classification and evaluation of results. The red element in the RGB model is selected as the threshold value parameter. The ripeness of pineapples is determined based on the percentage share of yellowish scales visible in images presenting the front and the back side of the fruit. The prototype system is capable of classifying pineap ples into three main groups: unripe, ripe, and fully ripe. The accuracy of 86.05% has been achieved during experiments.

2

Development of the Kuz-Ram model to blasting in a limestone quarry

Tosun A., Konak G., Toprak T., Karakus D., Onur A. H.

Archives of Mining Sciences

|

2014

|

Vol. 59, no. 2

477--488

EN

It is very important to know the degree of disintegration beforehand in open quarry bench blasting in terms of blasting efficiency. Kuznetsov (1973), Cunningham (1987), and Ouchterlony (2005) have developed the Rosin-Rammler distribution function for estimating the degrees of fragmentation (Rosin & Rammler, 1933). However, models they developed do not give realistic results for blast surfaces where there are a lot of discontinuity characteristics with broken and fissured areas due to the difference in structural characteristics of the rocks. In this study, average fragmentative distribution of the heap formed as a result of several blasting tests in a quarry belonging to BATIÇİM have been separately determined by using a dimensional analysis program with Wipfrag image processing technique and Kuz-Ram estimation model. Average granular size correction has been carried out with the approach of accepting as fine-grain the areas that couldn’t be determined by means of image analysis programs and that were neglected in the analysis. Subsequently, the land coefficient in Kuz-Ram model was determined to be 0.0383 rather than 0.06 for the said quarry by taking the average dimension values determined by Wipfrag method as reference point.

PL

Kluczowym zagadnieniem jest znajomość stopnia rozdrobnienia materiału przed przystąpieniem do prac strzałowych w kamieniołomach, dla określenia skuteczności strzelania. Kuznetsov (1973), Cunninghma (1987) i Ouchterlony (2005) wyprowadzili dystrybuantę bazującą na równaniu Rosina- -Rammlera (1933) w celu estymacji stopnia rozdrobnienia materiału. Jednakże opracowane modele nie oddają rzeczywistych wyników gdy prace strzałowe prowadzone są na powierzchniach w których znajdują się liczne strefy nieciągłości, załamań oraz spękań wskutek różnic we właściwościach struktur skalnych. W pracy tej obliczono średnią wielkość fragmentów materiałów rozdrabnianych w trakcie trwania prac strzałowych w kamieniołomie należącym do przedsiębiorstwa BATICIM. Wielkości te zostały określone oddzielnie przy użyciu programu do analiz wymiarowych wykorzystującego techniki przetwarzania obrazów Wipfrag oraz model estymacyjny oparty na modelu Kuz-Ram. Przeprowadzono korektę ze względu na wielkość uziarnienia, w podejściu tym jako obszary drobnoziarniste ujęto te obszary które nie mogły zostać zanalizowane przy użyciu technik przetwarzania obrazów i zostały pominięte w analizach. Współczynnik terenowy dla danego kamieniołomu według modelu Kuz-Ram został obliczony jako 0.0383 a nie 0.06, przy wykorzystaniu jako wielkości odniesienia średnich rozmiarów skał obliczonych z zastosowaniem metody Wipfraga.

3

Investigating the Performance of Various Relaxation Processes on the Surface Regularity and Dimensional Properties of Plain Knitted Fabrics Using the Image Processing Technique

Sanjari H. R., Semnani D., Sheikhzadeh M.

Fibres & Textiles in Eastern Europe

|

2011

|

Vol. 19, no. 2 (85)

36-42

EN

Te main objective of this paper was to develop a more advanced fnishing treatment for the stitch shape regularity and dimensional properties of knitted fabrics. In our previous work, we presented a novel defnition of knitted fabric regularity. In the present study we utilised our pervious approach based on Radon transformation analysis to study the efect of diferent relaxation regimes on the above properties of plain fabrics. We applied various relaxation methods to plain fabrics to make the stitch shape closer to the ideal. Te rhythm of stitch shape improvment was evaluated by measuring the deviation of the direction angle of the stitches against the ideal state. Te results show that the average fabric irregularity index afer ultrasonic-chemical relaxation treatment is better than other relaxation method indices. Moreover the average fabric constant dimensional parameter (Ks) afer the above treatment is higher than values reported by pervious workers. Terefore we propose this relaxation treatment as a new relaxation method that is more efective with respect to plain fabric regularity and stability during usage.