• 한국멀티미디어학회논문지
• /
• 제25권10호
• /
• pp.1386-1394
• /
• 2022

As consumption of high-quality fruits increases and sales and packaging units become smaller, the demand for automatic fruit grading systems is increasing. Compared to other crops, the quality of fruit is determined by visual characteristics such as shape, color, and scratches, rather than just physical size and weight. Accordingly, this study presents a CNN model that can effectively extract and classify the visual features of fruits and a perceptron that classifies fruits using physical features, and proposes a stacking ensemble model that can effectively combine the classification results of these two neural networks. The experiments with AI Hub public data show that the stacking ensemble model is effective for grading fruits. However, the ensemble model does not always improve the performance of classifying all the fruit grading. So, it is necessary to adapt the model according to the kind of fruit.

• Journal of Biosystems Engineering
• /
• 제20권1호
• /
• pp.95-103
• /
• 1995

A series of study has been conducted to develop a multi-purpose fruit grader using a black ＆ white image processing system equipped with a 550 nm interference filter. A device and high performance algorithms were developed for sizing and color grading of Fuji apple in the previous study. In this study an emphasis was put on finding correlations between weights of several kinds of fruits and their area fractions(AF), and on compensating the blurring effect upon sizing and color grading by conveying speed of fruit. Also, the effect of orientation and direction of fruit on conveyor during image forming was analyzed to identify any difficulty (or utilizing an automatic fruit feeder. The results are summarized as follows. 1. The correlation coefficients(r) between the weights of fruits and their image sizes were 0.984~0.996 for apples, 0.983~0.990 for peachs, 0.995 for tomato, 0.986 for sweet persimmon and 0.970~0.993 for pears. 2. It was possible to grade fruits by color with the area weighted mean gray values(AWMGV) based on the mean gray valves of direct image and the compensated values of reflected image of a fruit, and also possible to sort fruits by size with AF. Accuracies in sizing and color grading ranged over 81.0% ~95.0% and 82.0% ~89.7% respectively as compared with results from sizing by electronic weight scale and grading by expert. 3. The blurring effect on the sizing and color grading depending on conveying speed was identified and regression equations were derived. 4. It was found that errors in sizing and coloring grading due to the change in direction and orientation of Fuji apple on the conveyor were not significant as far as the stem end of apple keeping upward.

• Journal of Biosystems Engineering
• /
• 제18권4호
• /
• pp.358-370
• /
• 1993

In Korea, fruit grading has been mainly done manually, and manual grading depends on human sense. Thus it is subjected to human error and is not always as consistent as would be desired. Therefore, a study on the development of fruit grader was initiated to improve the consistency of fruit grading. The sensitivity for fruit weight of the conventional spring type weight grader has a tendency to decrease by physical characteristics of spring which is used as a weight sensing unit. This study was carried out to develop weight measuring device for establishing the base of weight sensing unit of electronic weight grader. This device consists of a weight sensor using load cell, data acquisition system, and a microcomputer containing program to calculate fruit weight. The weight measuring device using load cell was developed to increase sensitivity of fruit weight. The result of this study showed that the weight sensing unit of electronic weight grader contributed to the improvement of performance of weight measuring device.

• 스마트미디어저널
• /
• 제7권4호
• /
• pp.37-43
• /
• 2018

Most of fruit quality classification has been done by time consuming, inaccurate and intensive manual labor. This study proposed an automated fruit grading system based on appearances and internal flavors. In this study, image processing technique and a weight checker were used to measure the value of appearance features and the near infrared spectroscopy analysis method was used to estimate the value of internal flavors. Additionally, I suggested 8x8x5x5 ANN based fruit quality classifier model to grade fruits quality. The proposed automated fruit quality classification system is expected to be very beneficial for many farms where heavy manual labor is usually needed for fruit quality classification.

• 식물병연구
• /
• 제14권3호
• /
• pp.205-209
• /
• 2008

배 만수 품종에서 발생되는 수확 후 과피 갈변 발생원인을 구명하기 위해 2년간 시험한 결과를 요약하면 다음과 같다. 배 만수 품종의 과피 갈변은 수확 후 선과시 발생되는 미세한 물리적 상처 부위에서 갈변이 발생되었다. 따라서 과피에 인위적 상처를 낸 후 품종간 과피 갈변 정도를 비교한 결과 신고 품종에 비해 만수 품종이 과피 갈변 발생정도가 심하여 품종간 차이를 나타내었다. 또한 만수 품종은 신고 품종에 비해 과피 조직이 얇았으며, 과피내황산화능이 높고 함량이 많을수록 과피 갈변 발생이 적었다. 결과적으로 만수 품종의 과피 갈변은 수확후 선과주에 발생하는 물리적인 미세상처 발생을 억제하기 위해 수확 후 봉지를 씌운채 선과하면 과피의 상처 발생감소로 갈변 발생이 현저히 억제되었다.

• 농업과학연구
• /
• 제37권1호
• /
• pp.7-12
• /
• 2010

A significant difference in table grape quality was found between harvest seasons, producers and cultivars. In general, 'Kyoho' grapes showed much greater difference in fruit quality than 'Campbell Early'. The ratio of 'Campbell Early' grapes with poor quality (below quality standard within grades), was higher in fruit harvested early in the season, mainly because of immature fruit harvest. In 'Kyoho', poor quality of fruit seemed to be derived from the deviation of cultural practice between producers. Major factors responsible for poor quality in both cultivars includes harvest of unripe cluster, poor sorting and grading, berry abscission, and poor coloration. It is recommended to introduce a new quality standards considering the market condition in 'Kyoho' or a fresh-cut technology of grape berries for niche market.

• Journal of Biosystems Engineering
• /
• 제17권4호
• /
• pp.363-369
• /
• 1992

The aim of this study was to examine the blurring effects on performance of the experimental fruit grader in grading Fuji apples by size and coloration of the whole surface of individual apples. The grader consisted of a black/white image prcessing system, one camera, and utilized the algorithm developed for high speed sorting in the previous study. The results are summarized as follows : 1. With the algorithm developed in the previous study, it took 0.27~0.33 second in analyzing the size and coloration of an apple, and relative errors were within 3% for size and 1.3% for coloration. 2. The effect of blurring increased linearly with the conveying speed of apple and showed more significant effect on detection of coloration than on determining of size. 3. Considering the blurring effect, capacity of the experimental fruit grader was estimated to 7,500 apples per hour.

• 한국농업기계학회:학술대회논문집
• /
• 한국농업기계학회 1993년도 Proceedings of International Conference for Agricultural Machinery and Process Engineering
• /
• pp.1292-1301
• /
• 1993

This study was intended to examine feasibility of sizing and color grading of Fuji apple with black/white image processing system , to develop a device with which the whole surface of an apple could be captured by one camera , to develop an algorithm for a high speed sorting , and to examine the effects of blurring on the performance of the experimental fruit grader.

• 한국식품저장유통학회지
• /
• 제7권2호
• /
• pp.155-159
• /
• 2000

Apple fruit grading is largely dependant on skin color degree. This work reports about the possibility of nondestructive assessment of apple fruit color using infrared(NIR) reflectance spectroscopy. NIR spectra of apple fruit were collected in wavelength range of 1100~2500nm using an InfraAlyzer 500C(Bran+Luebbe). Calibration as calculated by the standard analysis procedures MLR(multiple linear regression) and stepwise, was performed by allowing the IDAS software to select the best regression equations using raw spectra of sample. Color degree of apple skin was expressed as 2 factors, anthocyanin content by purification and a-value by colorimeter. A total of 90 fruits was used for the calibration set(54) and prediction set(36). For determining a-value, the calibration model composed 6 wavelengths(2076, 2120, 2276, 2488, 2072 and 1492nm) provided the highest accuracy : correlation coefficient is 0.913 and standard error of prediction is 4.94. But, the accuracy of prediction result for anthocyanin content determining was rather low(R of 0.761).

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제15권2호
• /
• pp.421-441
• /
• 2021

Fruit detection in orchards is one of the most crucial tasks for designing the visual system of an automated harvesting robot. It is the first and foremost tool employed for tasks such as sorting, grading, harvesting, disease control, and yield estimation, etc. Efficient visual systems are crucial for designing an automated robot. However, conventional fruit detection methods always a trade-off with accuracy, real-time response, and extensibility. Therefore, an improved method is proposed based on coarse-to-fine multitask cascaded convolutional networks (MTCNN) with three aspects to enable the practical application. First, the architecture of Fruit-MTCNN was improved to increase its power to discriminate between objects and their backgrounds. Then, with a few manual labels and operations, synthetic images and labels were generated to increase the diversity and the number of image samples. Further, through the online hard example mining (OHEM) strategy during training, the detector retrained hard examples. Finally, the improved detector was tested for its performance that proved superior in predicted accuracy and retaining good performances on portability with the low time cost. Based on performance, it was concluded that the detector could be applied practically in the actual orchard environment.