Professional Documents
Culture Documents
Received: 2 January 1997 / Revision received: 21 January 1998 / Accepted: 12 February 1998
Abstract Apical and axial shoot tips of sweet potato were 1992). However, this propagation method is not cost-effec-
cultured to produce somatic embryos that mature and de- tive because of high labor costs, the need for maintenance
velop into plants in basal nutrient medium. However, the and multiplication of virus-free stocks, the requirement for
lack of high regeneration efficiency is an impediment to large nurseries, and lack of storage facilities. Mass produc-
the use of somatic embryos to produce synthetic seeds. tion of sweet potato somatic embryos for use as synthetic
Conversion experiments with mature embryos over a seeds and subsequent direct seeding would significantly re-
20-day period revealed that 80–90% of the embryos formed duce production costs for this vegetatively propagated crop.
roots but only 40–50% formed shoots. Using computer vi- A fairly homogeneous population of competent somatic
sion and canonical or Fisher discriminant function (CDA) embryos with a uniform and high rate of conversion is re-
analysis along with conversion results, it was possible to quired for the production of synthetic seeds. However, this
correctly classify competent embryos 40–50% of the time is very difficult to achieve because somatic embryos as a
based on size features, 50–60% of the time based on shape rule are highly heterogeneous in form, size, and maturity.
features, and 55–60% of the time based on color features. A machine vision system that can identify biological ma-
Non-competent embryos were correctly classified terial (Harrell et al. 1992; Molto and Harrell 1992) can be
65–75%, 55–60%, and 70–75% of the time based on size, very useful in identifying and selecting competent somatic
shape, and color, respectively. These results can be used embryos (Harrell et al. 1994). Computer vision analysis has
effectively to identify and select competent embryos for been used to qualify and quantify somatic embryogenesis
improved regeneration efficiency. (Cazzulino et al. 1990 a; Grand d’Esnon et al. 1988). Caz-
zulino et al. (1990 b) developed a model for carrot somatic
Key words Somatic embryo · Ipomoea batatas · embryo growth by estimating the number and developmen-
Computer vision · Conversion tal stage of embryos in suspension. Automated recognition
solely based on morphological features of different devel-
opmental stages of somatic embryos was demonstrated in
birch (Hämäläinen et al. 1993). However, a comparison of
Introduction visual morphological observation with subsequent conver-
sion of selected embryos into plantlets was not demon-
Sweet potato, a valuable source of food and biomass, is con- strated.
ventionally propagated through stem cuttings (Cantliffe We have developed a machine vision analysis system to
identify competent and non-competent classes of somatic
Communicated by I. K. Vasil embryos in sweet potato. Size, shape and color-related
K. Padmanabhan · D. J. Cantliffe (½) 1 measurements were used to classify the embryos following
Department of Horticultural Sciences, University of Florida, discriminant analysis, and were compared with the conver-
P. O. Box 110690, Gainesville, FL 32611, USA sion results in order to test the performance of the system.
Fax: +1-352-392-6479
R. C. Harrell
704 NE Sixth Street, Gainesville, FL 32601, USA
J. Harrison Materials and methods
Institute of Food and Agricultural Sciences,
Department of Statistics, University of Florida, Embryogenic culture initiation and maintenance
Gainesville, FL 32611, USA
Present address: Apical and axial shoot tips of sweet potato were used to initiate em-
1
8633 Datapoint Drive, No. 211, Phase II, San Antonio, TX 78229, USA bryogenic cultures (Liu and Cantliffe 1984). Embryogenic callus was
682
gently broken with a sterile glass slide tip and sieved through 710- (3) The ‘classification’ set consisted of 151 embryos, which were
µm and 355-µm mesh filters. The callus collected on each sieve was subjected to conversion studies. The purpose of the ‘classification’
rinsed with 3% (wt/vol) sucrose solution. The resultant 355–710 µm set was to test the validity of the discriminant function (Hair et al.
fraction was plated onto callus proliferation medium containing mod- 1987).
ified Murashige and Skoog (1962) basal medium with 30 mM KCl,
10 µM 2,4-dichlorophenoxyacetic acid (2,4-D), 1 µM benzyl adenine
(BA), 3% (wt/vol) sucrose and 0.7% Phytagar (Gibco) (Chee et al.
1992). The cultures were maintained in the dark at 30 °C in an
8-week transfer cycle. Results
Conversion results
Embryo production and harvest
Torpedo- and cotyledonary-stage somatic embryos were harvested Most embryos (80%) produced roots after 20 days in con-
from 7-week-old callus and transferred individually onto embryo version medium, while only half of the embryos produced
conversion medium (basal medium with 1.6% wt/vol sucrose) in pe- both roots and shoots (Table 1). Shoot conversion (50.8%)
tri dishes (60×15 mm). occurred after 10–15 days. In about 5% of the embryos
which formed both roots and shoots, callus proliferation
Machine vision and data analysis also occurred near their root poles.
Classification set
Competent 44% 56% 61% 49% 56% 44%
Noncompetent 34% 66% 41% 59% 33% 67%
ple, the competent and non-competent classes of somatic in ‘original,’ ‘training’ and ‘classification’ sets, respec-
embryos), and there are quantitative measurements of sev- tively.
eral variables of individuals under each group (Cruz-Cas- Based on the shape features, competent embryos were
tillo et al. 1994). classified 78%, 48%, and 61%, and non-competent em-
CDA generates linear composites of canonical variates bryos were classified 77%, 57%, and 59% correctly in the
(CDFs or canonical discriminant functions) of quantitative ‘original,’ ‘training’ and ‘classification’ sets, respec-
variables of individual samples (embryos) and then it sep- tively (Table 3). For the color-related features, 78%, 55%,
arates embryos maximally into two or more groups of in- and 56% of competent, and 54%, 73%, and 67% of non-
dividuals, while keeping the within-group variations small competent embryos were correctly classified in the ‘orig−
(Cruz-Castillo et al. 1994; Hair et al. 1987). inal,’ ‘training’ and ‘classification’ sets, respectively (Ta-
Once the CDF values were established using the ble 3). These classification results suggested that the mis-
‘original’ or ‘analysis’ sample, the procedure was applied classification rate was lower in identification of non-com-
to two different sets of embryos, the ‘training’ and ‘clas- petent classes of somatic embryos using size, shape and
sification’ sets. This was done to validate the discriminant color features. Embryos were correctly classified about
function (J. Harrison, personal communication). The per- 60% of the time as competent and 60–75% of the time as
formances of the discriminant function in the ‘original,’ non-competent (Table 3) in terms of their subsequent con-
‘training’ and ‘classification’ sets are presented in a con- version into plantlets. No additional improvement re-
fusion or classification matrix. This matrix compared the sulted with additional color feature extractions of the cot-
classification of competent versus non-competent em- yledonary and radicular region of embryos (results not
bryos. shown).
The diagonal entries of the matrix represented the frac-
tion of embryos correctly classified by the analysis. The
off fraction of the diagonal in the matrix represented in-
correct classification. The discriminant function, in the Discussion
case of size-related features, correctly classified embryos
as competent 70%, 47%, and 44% in ‘original,’ ‘training’ Identification of embryos with the greatest potential of
and ‘classification’ sets, respectively (Table 3). The non- conversion into plantlets is the most important require-
competent embryos were classified 85%, 73%, and 66%, ment for the development of a “synthetic seed” system
684