Egg volume estimation based on image processing and computer vision

Amount Estimation Method for Food Intake Based on Color and Depth Images through Deep Learning

In this paper, we propose an amount estimation method for food intake based on both color and depth images. Two pairs of color and depth images are captured pre- and post-meals. The pre- and post-meal color images are employed to detect food types and food existence regions using Mask R-CNN. The post-meal color image is spatially transformed to match the food region locations between the pre- and post-meal color images. The same transformation is also performed on the post-meal depth image. The pixel values of the post-meal depth image are compensated to reflect 3D position changes caused by the image transformation. In both the pre- and post-meal depth images, a space volume for each food region is calculated by dividing the space between the food surfaces and the camera into multiple tetrahedra. The food intake amounts are estimated as the difference in space volumes calculated from the pre- and post-meal depth images. From the simulation results, we verify that the proposed method estimates the food intake amount with an error of up to 2.2%.

Single-View Measurement Method for Egg Size Based on Small-Batch Images

Egg size is a crucial indicator for consumer evaluation and quality grading. The main goal of this study is to measure eggs' major and minor axes based on deep learning and single-view metrology. In this paper, we designed an egg-carrying component to obtain the actual outline of eggs. The Segformer algorithm was used to segment egg images in small batches. This study proposes a single-view measurement method suitable for eggs. Experimental results verified that the Segformer could obtain high segmentation accuracy for egg images in small batches. The mean intersection over union of the segmentation model was 96.15%, and the mean pixel accuracy was 97.17%. The R-squared was 0.969 (for the long axis) and 0.926 (for the short axis), obtained through the egg single-view measurement method proposed in this paper.

On-line weight estimation of broiler carcass and cuts by a computer vision system

In a broiler carcass production conveyor system, inspection, monitoring, and grading carcass and cuts based on computer vision techniques are challenging due to cuts segmentation and ambient light conditions issues. This study presents a depth image-based broiler carcass weight prediction system. An Active Shape Model was developed to segment the carcass into 4 cuts (drumsticks, breasts, wings, and head and neck). Five regression models were developed based on the image features for each weight estimation (carcass and its cuts). The Bayesian-ANN model outperformed all other regression models at 0.9981 R² and 0.9847 R² in the whole carcass and head and neck weight estimation. The RBF-SVR model surpassed all the other drumstick, breast, and wings weight prediction models at 0.9129 R², 0.9352 R², and 0.9896 R², respectively. This proposed technique can be applied as a nondestructive, nonintrusive, and accurate on-line broiler carcass production system in the automation of chicken carcass and cuts weight estimation.

Weight and volume estimation of poultry and products based on computer vision systems: a review

The appearance, size, and weight of poultry meat and eggs are essential for production economics and vital in the poultry sector. These external characteristics influence their market price and consumers' preference and choice. With technological developments, there is an increase in the application and importance of vision systems in the agricultural sector. Computer vision has become a promising tool in the real-time automation of poultry weighing and processing systems. Owing to its noninvasive and nonintrusive nature and its capacity to present a wide range of information, computer vision systems can be applied in the size, mass, volume determination, and sorting and grading of poultry products. This review article gives a detailed summary of the current advances in measuring poultry products' external characteristics based on computer vision systems. An overview of computer vision systems is discussed and summarized. A comprehensive presentation of the application of computer vision-based systems for assessing poultry meat and eggs was provided, that is, weight and volume estimation, sorting, and classification. Finally, the challenges and potential future trends in size, weight, and volume estimation of poultry products are reported.

Non-Destructive Identification and Estimation of Granulation in Honey Pomelo Using Visible and Near-Infrared Transmittance Spectroscopy Combined with Machine Vision Technology

Granulation is a physiological disorder of juice sacs in citrus fruit, causing juice sacs to become hard and dry and resulting in decreased internal quality of citrus fruit. Honey pomelo is a thick-skinned citrus fruit, and it is difficult to identify the extent of granulation by observation of the outer peel and fruit shape. In this study, a rapid and non-destructive testing method using visible and near-infrared transmittance spectroscopy combined with machine vision technology was applied to identify and estimate granulation inside fruit. A total of 600 samples in different growth periods was harvested, and fruit were divided into five classes according to five granulation levels. Spectral data were obtained for two ranges of 400–1100 nm and 900–1700 nm by visible and near-infrared transmittance spectroscopy. In addition, chemometrics were used to measure the chemical changes of soluble solid content (SSC), titratable acidity (TA), and moisture content (MC) caused by different granulation levels. Machine vision technology can rapidly estimate the external characteristics of samples and measure the physical changes in mass and volume caused by different granulation levels. Compared with using a single or traditional methods, the predictive performances of multi-category classification models (PCA-SVM and PCA-GRNN) were significantly enhanced. In particular, the model accuracy rate (ARM) was 99% for PCA-GRNN, with classification accuracy (CA), classification sensitivity (CS), and classification specificity (CSP) of 0.9950, 0.9750, and 0.9934, respectively. The results showed that this method has great potential for the identification and estimation of granulation. Multi-source data fusion and application of a multi-category classification model with the smallest number of input layers and acceptable high predictive performances are proposed for on-line applications. This method can be effectively used on-line for the non-destructive detection of fruits with granulation.