Study of an object recognition algorithm

This project introduces an improvement of the vision capacity of the robot Robotino operating under ROS platform. A method for recognizing object class using binary features has been developed. The proposed method performs a binary classification of the descriptors of each training image to characterize the appearance of the object class. It presents the use of the binary descriptor based on the difference of gray intensity of the pixels in the image. It shows that binary features are suitable to represent object class in spite of the low resolution and the weak information concerning details of the object in the image. It also introduces the use of a boosting method (Adaboost) of feature selection al- lowing to eliminate redundancies and noise in order to improve the performance of the classifier. Finally, a kernel classifier SVM (Support Vector Machine) is trained with the available database and applied for predictions on new images. One possible future work is to establish a visual servo-control that is to say the reac- tion of the robot to the detection of the object.

Net phytoplankton and zooplankton in the New York Bight, January 1976 to February 1978, with comments on the effects of wind, Gulf Stream eddies, and slope water intrusions

Results are given of monthly net phytoplankton and zooplankton sampling from a 10 m depth in shelf, slope, and Gulf Stream eddy water along a transect running southeastward from Ambrose Light, New York, in 1976, 1977, and early 1978. Plankton abundance and temperature at 10 m and sea surface salinity at each station are listed. The effects of atmospheric forcing and Gulf Stream eddies on plankton distribution and abundance arc discussed. The frequency of Gulf Stream eddy passage through the New York Bight corresponded with the frequency of tropical-subtropical net phytoplankton in the samples. Gulf Stream eddies injected tropical-subtropical zooplankton onto the shelf and removed shelfwater and its entrained zooplankton. Wind-induced offshore Ekman transport corresponded generally with the unusual timing of two net phytoplankton maxima. Midsummer net phytoplankton maxima were recorded following the passage of Hurricane Belle (August 1976) and a cold front (July 1977). Tropical-subtropical zooplankton which had been injected onto the outer shelf by Gulf Stream eddies were moved to the inner shelf by a wind-induced current moving up the Hudson Shelf Valley. (PDF file contains 47 pages.)

On the Selection of Non-Invasive Methods Based on Speech Analysis Oriented to Automatic Alzheimer Disease Diagnosis

The work presented here is part of a larger study to identify novel technologies and biomarkers for early Alzheimer disease (AD) detection and it focuses on evaluating the suitability of a new approach for early AD diagnosis by non-invasive methods. The purpose is to examine in a pilot study the potential of applying intelligent algorithms to speech features obtained from suspected patients in order to contribute to the improvement of diagnosis of AD and its degree of severity. In this sense, Artificial Neural Networks (ANN) have been used for the automatic classification of the two classes (AD and control subjects). Two human issues have been analyzed for feature selection: Spontaneous Speech and Emotional Response. Not only linear features but also non-linear ones, such as Fractal Dimension, have been explored. The approach is non invasive, low cost and without any side effects. Obtained experimental results were very satisfactory and promising for early diagnosis and classification of AD patients.

Fishing gear survey of Lake Alau

This report briefly documents the fishing gear commonly used by local fishers in Lake Alau (Nigeria). Five types of gears are described: gill-net, cast net, Gura trap, longlines, and Clap net

Fisheries statistics of Kainji Lake, northern Nigeria, Nov. 1994 - Dec. 1998

Results are presented for the first 4 years data (1994-1998) of the Kainji Lake catch assessment survey, collected and analysed by the Nigerian-German Kainji Lake Fisheries Promotion Project. The following areas are covered: Methodology and alterations of the original sampling concept; Frame survey results - factors relating to the CAS; Catch assessment survey results - general; Gill net fishery; Drift net fishery; Beach seine fishery; Cast net fishery; Longline fishery; Trap fishery; and, Catch statistics from fisherwomen. (PDF contains 143 pages)

Classifying Ecosystem Services for Economic Valuation: The case of forest water services

30 p.

Contributions to time series data mining departing from the problem of road travel time modeling

194 p.

Growth and emigration of white shimp, Litopenaeus vannamei, in the Mar Muerto Lagoon, Southern Mexico

Microcohorts of white shrimp, Litopenaeus vannamei, were sampled with a cast net at fortnightly intervals in the Mar Muerto Lagoon, Southern Mexico. Shrimp recruited to the lagoon throughout the sampling period (January to August 1993). Mean growth rates of microcohorts ranged from 0.21 to 1.21 mm total length (TL) per day. Juvenile shrimp mainly between the sizes of 70 to 80 mm TL emigrated from the lagoon. Growth and the onset of emigration appeared to be related to water salinity.

Study on fishing gears, species selectivity toward gears and catch composition of BSKB beel, Khulna, Bangladesh

An investigation on the types of fishing gear used and their species selectivity and effects on fishes of BSKB beel in Khulna was conducted from June '95 to January '96. Fishermen were found to follow 6 fishing techniques viz., netting, trapping, angling, spearing, dewatering and hand picking. Among them 23 types of the fishing gear was recorded to be used by the fishermen of which 7, 8, 4 and 4 are nets, traps, hooks and lines, and hand harpoon respectively. A total of 47 species of fish were identified in the catches of different gears used by the fishermen in BSKB beel. Particulars, mode of operation, fishing season and catch composition of different fishing gears were determined. Seine, cast and lift net, traps (charo, arinda and ghuni), and hooks and lines (dhawn and nol broshi) were recorded as nonselective gear considering the fish species caught. However, gill nets (punti, koi and fash jal), clasp nets (bhuti jal), some traps (khadom, tubo), hooks and lines (chip borshi, chasra) and all spears were used as more or less selective gear. With respect to species and its size fash jal, bhuti jal, trap (khadom, ramani), and koach, juti and jhupi among spears were regarded to be more or less large-species-gear. But punti jal, koi jal, trap (koi dughair, charo, tubo, arinda and ghuni), nol borshi and spear (ful-kuchi) were small-species-gear. Among all gears seine net, cast net, lift net, koi dughair and ramani were recorded deleterious for carps especially for stocked fingerlings. For relatively small sized wild fishes koi jal, punti jal and ghuni traps were identified as detrimental gear.

Present status of harvesting, transportation and marketing of freshwater Small Indigenous Species of Fish (SIS) of Bangladesh

A survey conducted to evaluate the present status of harvesting, transportation and marketing of freshwater SIS of Bangladesh showed that among the 56 to 73 species of small indigenous fish species (SIS), mola (Amblypharyngodon mola), dhela (Osteobrama cotio cotio), puti (Barbodes sarana), tit puti (Puntius ticto), tengra (Mystus vitatus), chapila ( Gudusia chapra), batashi (Pseudotropius atherinoides), buzuri tengra (Mystus tengra), katchki (Corika soborna), gol chanda (Chanda ranga), lamba chanda (Chanda nama), phul chela (Chela phulo) and lata (Channa punctatus) are the most common. But the abundance of the mola, puti, tengra, batashi and chapila are higher than the other SIS species. Puti was at the top of the list in percentage abundance (7-9%), followed by mola (7-8%), tengra 6-7% and chapila 5-6% in all the investigated areas. Three levels of market or marketing systems were observed in the distribution channel of fish trade: primary, secondary/higher secondary and final consuming markets. Primary market is the catching point in the rural area. In secondary market, the collectors bring the fish from primary market to the landing ghats, usually to the nearest thana market or at a place well linked by rivers, road and rail transport. On purchasing the fish from the higher secondary market, the paikars sell the fish to the retailers who take it to the final consumer market. Different types of crafts and gears are used for catching the SIS in Bangladesh. Monofilament gill nets (20%), traps and line nets (25%) and seine nets (25%) are the dominant gears in respect of the total catch followed by lift net (10%) and cast net (20%). A total of 10 types of gear are listed according to their mode of operation which used for catching SIS. The processors in consumer market received higher marketing profit followed by primary market and secondary market, respectively.

Some socio-economic aspects of the fishermen of twin pronged floodplain wetlands in West Bengal

A socio-economic investigation was carried out in two fishermen cooperative societies namely Purba Helatala Fishermen Co-operative Society (E-1), Barhal Fishermen Co-operative Society (E-2), under Maldah district, West Bengal to which the beels (flood plains) under study belong. A total of 132 member fishermen, which constituted the sample, were personally interviewed. The age group of the fishermen of the sample in E-1 varied between 20 and 66 years whereas in E-2 it was 22 and 61 years. All the members of the sample belonged to Scheduled Caste (SC) community. The primary occupation of all the respondents of both the beels was observed to be fishing (100%). Maximum number of illiterate respondents was observed to 56% in E-2 and 35% in E-1. It has been observed that as many as 38.3% of fishermen were having fishing experience which ranging from 16 to 20 years in E-1 whereas it was 6 - 10 years (36.1%) in E-2. Maximum number of fishermen lived in thatched houses (41.66%) in E-1 whereas in E-2 most of them lived in houses made of corrugated tin/tile shed (41.66%). As many as 41.55% of E-1 and 30.55% of E-2 used dug-out canoes for their fishing. Maximum number of fishermen used cast net with individualistic approach (100%) followed by Gill net (E-1:41.56% and E-2:55.55%). Most of the fishermen of the sample participated in fishing activities for 241 to 270 days (41.66%) in E-2 whereas it was 211 to 240 days (33.33 %) in E-1 in a year. During fishing season as many as 40.0% of the respondents of E-1 earned on an average Rs. 801.00 to Rs. 900.00 per month whereas it was Rs. 901.00 to Rs.1,000.00 (43.05%) in case of E-2. A section of fishermen of the sample borrowed money often (51.6%) E-1 whereas it was most often (27.27%) of E-2. The respondents of E-2 made regular repayment of the loan to the maximum extent (79.48%) whereas it was 57.44% in E-1. Higher fish production vis-a-vis higher income for the fishermen was observed in the beel (E2) having close characteristic.

Eradication of predatory fishes in fish ponds and nursery ponds using electricity

Prior to stocking economic species in tanks and nursery ponds, all the unwanted and predatory fishes in them will have to be removed. Even though most of them can be caught with ordinary gear such as cast net, drag net, scoop net etc. Total or absolute eradication is not possible and those left over are sufficient to destroy all the fingerlings introduced. Development of an electrical method for killing predatory fishes at reasonable cost is therefore, expected to provide a solution. This paper deals with a few methods, based on the use of electricity and which are suitable for different ponds.

Preliminary observations in the design and operation of a three panel, double trawl net

Fishing operations were conducted off Kakinada using a 3 panel double trawl net with twin cod-ends to study the utility of the net in catching both bottom and off bottom fishes. The observations indicate that the net is effective in simultaneous catching of bottom as well as off-bottom fishes and separating them while in operation. The design details of the net and the particulars of the fishing operations conducted are presented.

The study of biological characteristic and population dynamic of Caspian vimba (Vimba vimba persa), in coastal waters of Caspian Sea (Mazandaran Province)

This study was conducted to determine reproduction characteristics, diet regime, age structure and population dynamics parameters of the vimba vimba persa (Pallas, 1811) in Mazandaran waters of the Caspian Sea, from October 2008 to September 2009. A total of 994 specimens were monthly collected by beach seine and cast net from six fish landings of Ramsar, Tonekabon, Chaloos, Mahmood Abad, Sari and Behshahr. Biometric characters were measured for each specimen at the laboratory. Scales were used for age determination. Sex determination and fecundity were determined. Population dynamic parameters as well as stock assessment including cohort analysis were estimated using FISAT software. The finding showed that the mean of fork length and body weight of the Caspian Vimba were 168.4±2.6 mm and 71.94±32.24 g respectively. Strong correlation was found between these two variables (a= 0.012; b = 3.047; r2 = 0.955). 92 specimens were studied from the fecundity point of view. This species was found to have more abundance in spring (esp. Apr-May). The samples composed of 397(42.6%) male, 537(57.4%) female; Overall sex ratio (M: F =1: 1.35) was significantly different from the expected 1:1 ratio (p ≤0.05). The advanced stages of maturity (4th & 5th) were found in April and May. The highest Gonadosomatic Index in female was in May and the lowest one was in July. This fish is therefore a spring spawner. The maximum absolute and relative fecundities were 34640 and 260.9, respectively; the minimum absolute and relative fecundities were 5400 and 94.5 respectively. The averages of absolute and relative fecundities were 17198±7710 and 171.85±48.8, respectively. Coefficient vacuity index was 59.2% which indicates that this fish is mesophagous. Among of living creature consumes by Caspian Vimba mollusks, 76 arthropods, worms, plants, detritus and fishes were found 32.9% , 26.7% , 13.4% , 17% , 4.4% and 1.6% respectively. The infinite fork lengths were 261 mm for females, 25mm for males and 261 mm for both sexes respectively. For population growth and mortality parameters; K ( 0.28 per year for both sexes, 0.3 per year for males, 0.33 per year for females); t0 ( -0.65 year for both sexes, -0.23 year in females, -0.51 year in males ); Φ' ( 2.28 ); Z ( 0.98 per year ); M ( 0.59 per year); F ( 0.39 per year) and Exploitation coefficient was 0.4. The analysis showed that total biomass and MSY were 1336 and 528.8 tonnes respectively.

An Automatic Classification Method of Metabolic Cycle Based on Biomimetic Pattern Recognition

Correct classification of different metabolic cycle stages to identification cell cycle is significant in both human development and clinical diagnostics. However, it has no perfect method has been reached in classification of metabolic cycle yet. This paper exploringly puts forward an automatic classification method of metabolic cycle based on Biomimetic pattern recognition (BPR). As to the three phases of yeast metabolic cycle, the correct classification rate reaches 90%, 100% and 100% respectively.