Rapid detection and visualization of slight bruise on apples using hyperspectral imaging

Abstract

During harvesting and storage, slight bruises on apple surface caused by impact, compression, vibration, or abrasion are inevitable. To find an appropriate method to identify the bruised apples at five stages (1 min, 1 day, 2 days, 3 days and 4 days after bruising), 108 Fuji apples were collected as samples. Hyperspectral images of apples covering the wavelength between 400 and 1000 nm were acquired by the SOC710-VP hyperspectral imaging system. The standard normal variate (SNV) method was utilized for smoothing and denoising of the original hyperspectral data. Classification models, including Extreme Learning Machine (ELM), Partial Least Squares Linear Discriminant Analysis (PLS-DA) and Classification and Regression Tree (CART), coupled with a variable selection method named Minimum Redundancy Maximum Relevance (mRMR), were built to identify the bruised apples. The results showed that the ELM models exhibited the best classification capability, with the mean correct classification rate of 95.97%. The bruised samples are easier to be identified over time. Minimum noise fraction (MNF) method was implemented to classify the bruised region of apples based on the selected wavelengths. The overall classification accuracy of MNF is 92.9%, which indicates that MNF is an effective method for identifying bruised regions of apples. ARTICLE HISTORY

Early detection of slight bruises in apples by cost-efficient near-infrared imaging

“…Therefore, the impact energy (i.e., the kinetic energy) could be adjusted by changing the releasing angle of the pendulum arm. In the drop test, bruises could be generated by dropping the fruit to a contact surface [13], [23], [29]- [32] or by releasing a small steel ball to the fruit from a specific height [26], [33]- [36]. Because the impact energy computed as the potential energy that depends on the weight of the dropped object, a consistent impact level could be achieved for weight-varying apples by dropping a specific steel ball to the apples.…”

“…Detecting bruises at the early stages is an interesting postharvest problem to reduce the cost incurred in transportation, storage, and shortened shelf-life. Early detection approaches were evaluated with apples immediately after being subjected to a bruising experiment [7], [10], and [23]. However, few studies focused on slight bruises with well-defined impact levels, partly due to no standard method currently available for assessing the severity of fruit bruising [24].…”

“…However, few studies focused on slight bruises with well-defined impact levels, partly due to no standard method currently available for assessing the severity of fruit bruising [24]. A controlled bruise was considered a slight bruise if it was difficult to recognize based on its texture and color [23]. Such definition was subjective and might result in inconsistent classification of impact levels.…”

Thoughts for Foods: Imaging Technology Opportunities for Monitoring and Measuring Food Quality

“…Also, HSI systems were applied for the ripeness monitoring of a large number of different fruit varieties [141][142][143][144][145][146][147]. Also, defects or blemishes detection such as bruising in fruit [147][148][149][150][151][152][153][154][155]. Recent studies on 3) Binary mask selecting the center of the cake (white), air bubbles (black), and ( 4) cake image ready for spectral data extraction.…”

Classification of Robust and Rotten Apples by Deep Learning Algorithm

“…Gerçekleştirilen çalışmada, CAPA veri tabanından alınan görüntülerin derin öğrenme algoritması kullanılarak, elmaların çürük ve sağlam olarak sınıflandırılması sağlanmıştır.2. Materyal ve MetotÇalışmada, Gembloux Tarım Üniversitesinde bir araştırma sonucunda, Literatürde son yıllarda yapılan, elma görüntülerinden sağlam ve çürük olarak sınıflandırma ile ilgili yapılan çalışmalar incelendiğinde, hiperspektral ve multispektral görüntüleme sistemlerinin kullanıldığı görülmektedir[10][11][12][13][14][15][16]. Fakat tercih edilen sistemlerin genellikle laboratuvar ortamında çalışabilecek nitelikte olduğu ve maliyetlerinin ise çok fazla olduğu görülmektedir.…”