Koichiro Enomoto’s research while affiliated with Shiga University and other places

In scallop aquaculture, it is important to monitor the growth of scallops in the sea for growth control. In this paper, we propose a method to automatically detect scallops and determine whether they are alive or dead based on the information obtained by 3D reconstruction of scallops in the underwater from images. The scallops are detected using Sparse ICP, and multiple initial positions are set to deal with the problem of local solutions. The scallops are judged survival by defining an opening angle from the obtained scallop shell positions. The experimental results showed that the scallop detection rate was 100% for the simulation data and 81.3% for the real data. The survival rate was 76.9% for the real data. These results show that the number of samplings is three times larger than that of the previous method, showing the effectiveness of the proposed method.

The current scallop fishing has some problems. The fishing can not harvest the natural scallops for raw consumption by the problems. To solve the problems, this study proposes a new method using a robot arm. The new method needs a gripper attaching a robot arm. From some demands that meet for scallops gripping, this research judge a suction gripper gripping with negative pressure is suitable to harvest the scallops. However, a suction gripper has disadvantages for gripping the scallops. Therefore, we develop a new gripper with a built-in spring to improve the disadvantages. Additionally, this study experiments with the developed gripper and evaluates the developed gripper.

In this paper, we propose three methods for removing floating particles from underwater images. The first two methods are based on Generative Adversarial Networks (GANs). The first method uses CycleGAN which can be trained with an unpaired dataset, and the second method uses pix2pixHD that is trained with a paired dataset created by adding artificial particles to underwater images. The third method consists of two-step process – particle detection and image inpainting. For particle detection, an image segmentation neural network U-Net is trained by using underwater images added with artificial particles. Using the output of U-Net, the particle regions are repaired by an image inpainting network Partial Convolutions. The experimental results showed that the methods using GANs were able to remove floating particles, but the resolution became lower than that of the original images. On the other hand, the results of the method using U-Net and Partial Convolutions showed that it is capable of accurate detection and removal of floating particles without loss of resolution.

We aimed to establish an objective method to evaluate the quality of dried sea cucumber products. Clearness of papillae on the body wall affects the quality and commercial value of this luxury seafood. Unfortunately, an objective method for evaluating papilla clearness remains to be developed. We digitalized clearness of papillae from dried sea cucumber products images using TouchDeMeasure. This software application was developed by the authors. Using this tool, we measured body length (BL), total perimeter including all papillae (TP), and ellipse perimeter excluding the papillae (EP). Results showed that TP/BL, TP/EP, and (TP-EP)/BL values for a clear papilla were significantly higher than those for an unclear papilla. Additionally, we confirmed that these parameters corresponded with the grading levels used by a connoisseur. These experimental results show that our method digitally analyzes the clearness of papillae to evaluate the quality of dried sea cucumber products.

The current method of scallop fishing is mainly Scallop Dredge. When sand at sea bottom enters a scallop during harvesting, none of the scallops can be sold for raw. Additionally, Scallop Dredge harvests a scallop with marine sources such as other shells. To avoid these problems, this study proposes a new harvesting method using a robot arm. The new method can catch the only scallop without entering sand in a scallop. the new method needs a gripper to attach a gripper to a robot arm. This study focuses on the phenomenon of adsorption by negative pressure and tries to develop a negative pressure adsorption gripper for scallop harvesting. an experiment gripping a scallop with the gripper of this study was successful.

In this paper, we propose a method for eliminating image noise (from sand, algae, bubbles, etc.) that is transferred when capturing underwater images. In detail, we separate the input of a moving underwater image into a slowly changing background and foreground moving at a certain speed using Kalman filter and object tracking. In this research, we examined the difference in image quality of the output image by changing the noise variance given to the Kalman filter.

Trade‐offs are considered key to understanding mechanisms supporting the coexistence of multiple plant species. Thus, understanding the mechanisms underlying trade‐offs is expected to contribute to conservation and management of macroalgal beds composed of diverse macroalgae of rocky shore ecosystems. To test the occurrence of trade‐offs between productivity and thallus toughness as well as pair‐wise thallus trait relationships that are expected to indirectly relate to any trade‐offs, traits and relationships for 13 species of macroalgae from the central area along the Japan Sea coast of Honshu, Japan were examined. In each species we examined for photosynthetic capacity per unit biomass (as Amass) and nitrogen (i.e., photosynthetic nitrogen‐use efficiency, PNUE), nitrogen content (as Nmass), thallus mass per unit thallus area (as TMA) and force required to penetrate the thallus (as Fp, a common index of leaf toughness in land plants by punch test). A significant negative correlation indicating a trade‐off between productivity and thallus toughness was found between Amass or PNUE and Fp. Pair‐wise relationships that were expected to indirectly relate to the trade‐off were as follows. Amass was positively correlated with Nmass. Thalli with high Nmass extensively utilizing nitrogen in the photosynthetic parts, and consequently exhibiting elevated metabolic rates. Moreover, thalli with high Nmass tended to be associated with low TMA, and Nmass decreased with increasing TMA. A significant negative correlation was observed between TMA and Amass or PNUE because of the linkage of high Amass or PNUE with high Nmass and high Nmass associated with low TMA, while a significant positive correlation was observed between TMA and Fp. The two correlations indicate a physiological and structural trade‐off, which underlies the interdependency of thallus traits. Results of multivariate analyses also indicated that the thallus traits interdependently vary across a single axis based on the trade‐off.

In this study, we propose a bottom-sediment classification system using seabed images. Seabed videos using a digital video (DV) camera were taken for a fishery-resource survey in the scatter scallop fishing grounds in Hokkaido, Japan. Scatter scallop fishing is a method of waiting for naturally growing young shells on the seabed. We acquired about 0.02 km ² of seabed videos in 2015 in Monbetsu. We cannot survey as wide a range using the DV camera as we can using sonar; however, we can obtain high-resolution 75 × 42 cm ² seabed images. We can classify bottom sediment in a narrower range than bottom-sediment classification methods using sonar. Our research aims to classify the following four types of bottom sediment: sand, ballast, gravel, and shell bank. The bottom sediment affects the growth of scallops and the survival rate of young shells. Therefore, understanding the undersea environment is important. In this study, we used a convolutional neural network (CNN) for the bottom-sediment classification from seabed images. Using CNN enables automatic and high-speed classification. This experiment showed average accuracies of about 95% for three types of bottom sediment and 76.5% for the fourth type (gravel). Moreover, we created a fishing-ground map based on the bottom sediment for visualizing the seabed environment.

In farming Japanese scallops Mizuhopecten yessoensis, larvae investigation and efficient seed collection are very important processes. During yearly larvae investigation from May to June, fishery managers count the number of larvae and measure the shell size. Identification technology has been developed using immunostaining to mark scallops. The resulting fluorescence images contain both fluorescence-stained and autofluorescent larvae. However, there is currently no technology for evaluating these images automatically, so the experts have to perform these tasks manually. Our overall aim is to develop a system for automatically measuring the number and size of scallops. In this paper, we propose a method for detecting fluorescence-stained larva and measuring the items in the fluorescence images. We will present the experimental results and discuss the effectiveness of the proposed method. We will also introduce an application that facilitates measurement of scallops.